EU Kids Online Rep Cover_3_AD.indd

Transkript

Co-funded by the European Union
%5+IDS/NLINE
Comparing children's online opportunities
and risks across Europe
European Research on Cultural, Contextual and Risk Issues
in Children’s Safe Use of the Internet and New Media (2006-2009)
A project funded by the EC Safer Internet Plus Programme
– http://ec.europa.eu/saferinternet
ww
www.eukidsonline.net
www
ww.e
Comparing children's online
opportunities and risks across Europe:
Cross-national comparisons for EU Kids Online
Editors
Uwe Hasebrink, Sonia Livingstone and Leslie Haddon
Contributors
Verónica Donoso, Cédric Fluckiger, Carmelo Garitaonandia,
Maialen Garmendia, Jos de Haan, Leslie Haddon, Uwe Hasebrink,
Lucyna Kirwil, Yiannis Laouris, Benoit Lelong, Sonia Livingstone,
Bojana Lobe, Jivka Marinova, Helen McQuillan, Kjartan Olafsson,
Pille Pruulmann-Vengerfeldt, Katia Segers, José Alberto Simões,
Vaclav Stetka, Liza Tsaliki, Anna Van Cauwenberge,
Thomas Wold.
This is a report from the EU Kids Online network, June 2008
For a complete list of participants, see Annex B
Please cite this report as follows:
Hasebrink, U., Livingstone, S., Haddon, L. (2008) Comparing children’s online
opportunities and risks across Europe: Cross-national comparisons for EU Kids
Online. London: EU Kids Online (Deliverable D3.2).
European Research on Cultural, Contextual and Risk Issues
in Children's Safe Use of the Internet and New Media
The EU Kids Online network is funded by the EC Safer Internet plus Programme) from 2006-9
(SIP-2005-MD-038229; http://ec.europa.eu/information_society/activities/sip/index_en.htm). It
examines research findings from 21 member states into how children and young people use
the internet and new online technologies. This three year collaboration aims to identify
comparable findings across Europe and to evaluate the social, cultural and regulatory
influences affecting online opportunities and risks, along with children’s and parents'
responses, in order to inform policy. It is charting available data, pinpointing gaps and
identifying factors that shape the research capability of European research institutions. Finally,
it examines methodological issues relating to the cross-cultural study of children’s online
experience. See www.eukidsonline.net.
2
Contents
1. Introduction .............................................................................................................................. 5
1.1 European research on children’s online activities and risks ........................................... 5
1.2 Why a comparative approach? ......................................................................................... 5
1.3. The EU Kids Online network ............................................................................................ 6
1.4 Scoping online risks and opportunities............................................................................. 6
1.5 Structure of the research field ........................................................................................... 6
1.6 Classifying risks and opportunities ................................................................................... 8
1.7 Varieties of cross-national comparison .......................................................................... 11
1.8 The organisation of this report ........................................................................................ 12
2. Children and young people’s internet use: Comparing across countries....................... 13
2.1 Access and usage............................................................................................................ 13
2.1.1 Parents’ and children’s internet use ....................................................................... 13
2.1.2 Locations of children’s internet use ........................................................................ 17
2.1.3 Positional factors influencing children’s internet use ............................................ 20
2.2. Risks and opportunities .................................................................................................. 24
2.2.1 Opportunities experienced by children online ....................................................... 25
2.2.2 Risks experienced by children online..................................................................... 28
2.2.3 Positional factors influencing children’s online opportunities and risks ............... 32
2.2.4 Consequences of online risks and coping ............................................................. 42
2.2.5 Relation between online opportunities and risks................................................... 46
2.3. Attitudes and skills .......................................................................................................... 47
2.4. Mediation by parents, teachers and peers.................................................................... 50
2.5 Conclusions ...................................................................................................................... 61
2.5.1 Conclusions regarding the general theoretical model........................................... 61
2.5.2 Conclusions regarding the classification of countries ........................................... 71
3. Explaining differences and commonalities between countries............................................ 77
3.1 Media Environment .......................................................................................................... 78
3.1.1 Internet and broadband diffusion............................................................................ 78
3.1.2 Internet safety tools ................................................................................................. 79
3.1.3 Media content for children....................................................................................... 82
3.2 Internet regulation and promotion................................................................................... 84
3.2.1 Legislation and Policing (Regulation)..................................................................... 84
3.2.2 The role of government and regulator.................................................................... 87
3.2.3 The influence of NGOs............................................................................................ 89
3.3 Public discourses ............................................................................................................. 90
3.3.1 Media coverage ....................................................................................................... 90
3.3.2 Role of NGOs and related stakeholders in shaping public discourses................ 92
3.3.3 Key Events ............................................................................................................... 93
3.4 Values and attitudes ........................................................................................................ 94
3.4.1 Sources of information ............................................................................................ 94
3.4.2 Commonalities and differences between the countries ........................................ 94
3.4.3 Indicators for classifications of the European countries ....................................... 94
3.4.4 Hypotheses regarding the influence of this contextual factor on Safer Internet . 96
3.5 Educational system.......................................................................................................... 96
3.5.1 General literacy of the population........................................................................... 96
3.5.2 The education of the parents’ generation .............................................................. 98
3.5.3 The kind of education for today’s children ........................................................... 100
3.5.4 The technical infrastructure of schools ................................................................ 102
3.5.5 Internet and media education ............................................................................... 104
3.6 Background factors ........................................................................................................ 106
3.6.1 Levels of social change ......................................................................................... 106
3.6.2 Inequalities ............................................................................................................. 107
3.6.3 Urbanisation ........................................................................................................... 108
3.6.4 Work and social class............................................................................................ 109
3.6.5 Free speech and censorship ................................................................................ 110
3.6.6 Migration and cultural homogeneity ..................................................................... 110
3.6.7 Role of the state..................................................................................................... 111
3.6.8 Language ............................................................................................................... 112
3
3.6.9 ‘Bedroom culture’................................................................................................... 113
Conclusions .......................................................................................................................... 115
4.1 Overview......................................................................................................................... 115
4.2 Summary of findings from the comparative analysis................................................... 116
4.3 Classification of countries in terms of children’s online risk........................................ 118
4.4 Contextual explanations for cross-national differences .............................................. 119
4.5 Commentary on the comparative process ................................................................... 123
5. Bibliography .......................................................................................................................... 126
6. Annexes ................................................................................................................................ 129
Annex A: EU Kids Online ....................................................................................................... 129
Annex B: EU Kids Online members....................................................................................... 130
4.
4
1. Introduction
1.1 European research on children’s online activities and risks
In many countries, within and outside Europe, children and young people are gaining access to
the Internet and online technologies at a rapid pace. As a Eurobarometer survey on Safer
Internet issues showed (EC, 2006), half of all children under 18 years old in the EU25 have
used the Internet, with even higher figures applying to teenagers. However, there are
substantial differences across countries (ranging from less than a third of children in Greece
and Bulgaria to over two thirds of those in Estonia and Denmark).
To understand what these changes mean for children and their families, for their education,
leisure, participation and community and, more negatively, for the risk of harm to children and
young people, this growing use of the Internet and online technologies is being closely tracked
by empirical research. Research teams across Europe are conducting empirical studies of
varying range and depth, in order to advise policy-makers how best to maximise the benefits
and minimise the risks associated with the changing media environment.
The collective findings of multiple research projects must now be integrated and their diverse
insights brought into focus.
1.2 Why a comparative approach?
The EU Kids Online network is premised on the assumption that a cross-national perspective
is vital, for children’s experiences of online technologies may – or may not - differ in different
countries; after all, countries vary in terms of family structures, education systems, attitudes to
technology, media regulation, social values, and much more.
As each country seeks to balance the possible failure to minimise the dangers against the
equally problematic failure to maximise the opportunities, cultural factors come to the fore. For
example, protection of children is a universal value, yet in practice different countries – for
reasons of religion, family structure, market competitiveness and media history – regard new
online risks through a cultural lens, asserting their own priorities, often motivated by implicit
values. To take another example, it may be that the incidence of risk is higher in countries
where diffusion has come later, or where media literacy is lower.
Without a comparative perspective, national studies risk two fallacies – that of assuming one’s
own country is unique when it is not, and that of assuming one’s own country is like others
when it is not. Researchers and policy makers are faced with asking themselves, for example,
whether research conducted in Germany applicable in Italy or whether findings from Northern
1
Europe suggest lessons for new accession countries?
How should one avoid these fallacies? The body of available evidence raises crucial questions
regarding expectations for, and interpretation of findings. Do we expect the risks faced by
children in one country to be the same as those in another? What are the costs of assuming
pan-European similarities, as a matter of convenience or pragmatism, potentially
underestimating the importance of local contexts of use? As part of the wider effort of
researchers and policy makers who seek a shared knowledge – of risks, of contexts and of
local distinctiveness – this report presents a critical analysis of the ways in which European
countries resemble in other as regarding children’s online risk and safety, and where and why
they may differ.
1
Partly, this is a methodological matter, as explored in Work Package 4 (‘Methodological Issues’) of the
EU Kids Online network, for one must determine whether survey methods developed in, say, Sweden are
straightforwardly replicated in Belgium? Partly too, it is a matter of the availability of data, as examined in
Work Package 1 (‘Data Availability and Gaps’).
5
1.3 The EU Kids Online network
The EU Kids Online thematic network comprises research teams in each of 21 countries
across Europe, tasked with keeping track of recent and ongoing empirical studies. In order to
provide a bridge between the specialist domain of empirical research and the policy
imperatives of safer Internet initiatives, the EU Kids Online network is examining European
research (national and multi-national) on cultural, contextual and risk issues in children's safe
use of the Internet and new media. It focuses on the intersection of three domains:
!
Children (mainly up to 18 years old), their families, domestic users;
!
Online technologies: mainly but not only the Internet; focussing on use and risk;
!
European empirical research and policy, prioritising the 21 countries in the network.
Working closely together since June 2006, the 21 national teams that comprise EU Kids Online
have developed constructive working arrangements that capture diversity across member
states and facilitate the identification of common patterns, themes and best practice.
EU Kids Online outputs are the collective effort of the EU Kids Online network. Network
members meet several times per year and work in close contact electronically in between. The
editors then integrate contributions and produce the final text for each report.
For further information, see Annex A and www.eukidsonline.net.
1.4 Scoping online risks and opportunities
The focal concern of EU Kids Online is children’s online risks and opportunities – these are,
therefore, our main dependent variables. Exactly what ‘risks’ and ‘opportunities’ includes is a
moving target. But it may reasonably be scoped as follows:
Online opportunities
!
!
!
!
!
!
!
!
!
!
!
!
!
Online risks
!
!
!
!
!
!
!
!
!
!
!
!
!
Access to global information
Educational resources
Social networking for old/new friends
Entertainment, games and fun
User-generated content creation
Civic or political participation
Privacy for expression of identity
Community involvement/activism
Technological expertise and literacy
Career advancement or employment
Personal/health/sexual advice
Specialist groups and fan forums
Shared experiences with distant others
Illegal content
Paedophiles, grooming, strangers
Extreme or sexual violence
Other harmful or offensive content
Racist/hate material/activities
Advertising/commercial persuasion
Biased/misinformation (advice, health)
Exploitation of personal information
Cyber-bullying, stalking, harassment
Gambling, financial scams
Self-harm (suicide, anorexia, etc)
Invasions/abuse of privacy
Illegal activities (hacking, downloading)
However, children’s and young people’s access to and use of online technologies occurs within
a broader context – domestic, familial, social, cultural, political, economic, etc. Many factors
may potentially influence their use in general and the risks they may encounter in particular. To
organise the potentially vast array of factors, we have classified these factors as dependent,
independent, mediating and contextual variables, as explained below.
1.5 Structure of the research field
The experience of online opportunities and risks is expected to vary according to children’s age
and gender, as well as by the socioeconomic status (SES) of the household (or other
stratifying factors such as parental education or urban/rural location). These sociodemographic factors are the main independent variables to account for differences in
opportunities and risks, though others may arise as the research findings are examined.
6
These socio-demographic factors influence children’s Internet access, online usage, and their
related attitudes and skills. These latter may be considered mediating variables, for they are
influenced by socio-demographic factors and they may influence online opportunities and risks.
Additional mediating variables are introduced by the activities of others – parents, teachers and
peers. Parents mediate, or regulate, their children’s online activities, thereby potentially
influencing their experience of opportunities and risks. For teachers and peers, further
influences may be expected, though these have been little researched. Such mediating
processes may, in turn, be influenced by parents’ own Internet use, or teachers’ online skills, or
domestic practices of media regulation more broadly.
Finally, we note key contextual variables likely to affect children’s online experiences. These
national or macro-societal factors include a) the media environment, b) ICT regulation, c) the
public discourse on children’s Internet use and possible risks of the Internet, d) general values
and attitudes regarding education, childhood, and technology and e) the educational system.
A framework that includes each of these key variables is shown in Figure 1.1.
Figure 1.1: Overview of the structure of the research field
Online activities of children
Access
Age
Risks and
opportunities
Gender
Attitudes
and skills
Usage
SES/inequality
Mediation by parents, teachers and peers
Individual level of analysis
Media
environment
ICT
regulation
Public
discourse
Attitudes
and values
Educational
system
Country level of analysis
The above figure provides a heuristic device for categorizing the key variables and specifying
the hypothetical links among them. The research field is divided, first, into an individual (or
child-centred) level of analysis for examining patterns of similarity and difference within
countries; and second, into the country (or macro-societal) level of analysis for examining
2
patterns of similarity and difference across countries.
2
It is not our intention to focus on the individual child separated from their social context but rather to
show how children are located in a network of social influences at all levels from the familial to the
societal. Analytically, it is useful to distinguish intra-country comparisons, for which the individual child is
the unit of analysis, from inter-country comparisons, for which the country is the unit of analysis.
7
Thus, Figure 1.1 represents a working hypothesis by which observed similarities and
differences across countries in children’s experiences of online opportunity and risk may be
explained in terms of the key variables identified in the research literature.
1.6 Classifying risks and opportunities
To analyse actual experiences with online risks and opportunities throughout Europe, we must
bring together case studies on the national level. In these studies risks and opportunities are
defined quite heterogeneously. In order to relate these studies to each other a systematic
approach to the definition of Internet related risks has been developed.
The overall model is as follows (see Figure 1.2 for risks and Figure 1.3 for opportunities). Risks
and opportunities refer to negative or positive experiences that might happen. Negative or
positive experiences result from transactions between communicators, the content/services
they provide and the user. The two necessary conditions for these transactions are:
!
Access: This is the obligatory condition for any negative or positive experience related to
the Internet and so may be regarded, in itself, as either “risk” or “opportunity”. There will
be differences between various places or occasions where children have access, e.g. at
home, at school, with friends, which differ with respect to the degree of regulation or
guidance by parents, teachers etc.
!
Usage: Given access, the nature of children’s use of online media is also a crucial
condition of risk. The longer children use online media and the more they use certain
services, the more likely they are to encounter certain negative or positive experiences.
However, beyond children’s preferences for more or less risky online activities, factors
such as children’s online skills and media literacy may exacerbate or alleviate risks.
In what follows, the model is explained for risks (see Figure 1.2). An equivalent model for
opportunities is shown in Figure 1.3. The table on the top right side of Figure 1.2 classifies
different types of risks. The starting point was to ask, “What processes lead to different risks?”
The model assumes a transaction between communicative motivations and the role of the child
when going online. The row headings of the table refer to the forms of communicative roles:
!
Content – child as recipient (of mass communication)
!
Contact – child as participant (of peer/personal communication)
!
Conduct – child as actor (offering content or acting in personal contacts)
The column headings refer to motivations leading to risks – potentially problematic aspects of
the provision of particular contents and services online. Each cell provides examples for the
specific risk which arises from the transaction between the motivations and the child’s role.
In the lower part of the figure we note which negative consequences or effects might follow
from the four motivations and their transaction with the child’s behaviour. An additional area of
negative consequence is linked to usage: independent of risks which arise from negative
motivations, time consuming online activities (sometimes interpreted as Internet addiction) may
be negative consequences of Internet usage.
We have to note the limitations of this model:
!
Sometimes boundaries are blurred (e.g. aggression and sexuality can co-occur).
!
Issues of privacy and personal information cut across cells.
!
Some categories (e.g. sexuality) cover rather different kinds of risk.
Figure 1.3 provides an equivalent table for opportunities. The three rows of the table stay the
same, and four “positive motivations” have been defined - Education and Learning;
Participation and civic engagement; Creativity; Identity and social connection.
8
Access
Usage
At home or at
school or with
friends,
controlled by
whom?
Which kinds
of services for
which
purposes?
Frequency and
duration of
use
Time,
addiction,
dependency
Motives
Child‘s role
Commercial
interests
Aggression
Sexuality
Values/ ideology
Content
(What is found on
the web)
“Child as
recipient”
Advertising,
exploitation of
personal
information
Violent web
content
Problematic
sexual web
content
Biased
information,
racism,
blasphemy, health
“advice”
Contact
(Someone else
making contact)
“Child as
participant”
More
sophisticated
exploitation,
children being
tracked by
advertising
Being harassed,
stalked, bullied
Being groomed,
arranging for
offline contacts
Being supplied
with
misinformation
Conduct
(Child contacts
someone)
“Child as actor”
Illegal downloads,
sending offensive
messages to peers
Cyberbullying
someone else,
happy slapping
Publishing porn
Providing
misinformation
Financial
consequences,
privacy in commercial respect
Harm, anxiety,
aggressiveness
Personal contact
with strangers,
harm, sexual
orientation
Orientation to
aspects of real life
(health, politics,
etc)
Negative Consequences
Figure 1.2: Risks as transactional results of access, usage, the child‘s role, and underlying communicative motives leading to negative consequences
9
Access
Usage
At home or at
school or with
friends,
controlled by
whom?
Which kinds
of services for
which
purposes?
Frequency
and duration
of use
Technical
skills, media
literacy
Motives
Child‘s role
Education and
Learning
Participation
and civic
engagement
Creativity
Identity and
social connection
Content
(What is found on
the web)
“Child as
recipient”
Educational
resources
Global
information
Diversity of
resources
Advice (personal
/ health / sexual
etc)
Contact
(Someone else
making contact)
“Child as
participant”
Contact with
others who share
one’s interests
Exchange among
interest groups,
Being invited or
inspired to
participate in
creative processes
Social
networking,
shared
experiences with
distant others
Conduct
(Child contacts
someone)
“Child as actor”
Self-initiated and
collaborative
forms of learning
and education
Concrete forms of
civic engagement
User-generated
content creation
Expression of
identity
Knowledge, skills,
career
advancement
Civic
engagement
Creative skills
Identity and
social connection
Positive Consequences
Figure 1.3: Opportunities as transactional results of access, usage, the child‘s role, and underlying communicative motives leading to positive
consequences
10
1.7 Varieties of cross-national comparison
The core objective of the EU Kids Online network has been to systematically collect and compare findings
regarding online risks and opportunities in Europe. To realize this objective, the network built on systematic
considerations on comparative research elaborated by Kohn (Kohn, 1989) - see also Livingstone’s work
(Livingstone, 2003), who distinguished several types of comparative analysis.3
Type 1: Countries as objects in their own right
Treating countries as objects of analysis in their own right employs an idiographic lens. It aims to understand
particular countries for their own sake, with comparison representing a useful strategy for ‘seeing better’ and
so determining what is distinctive (or not) about a country (thus avoiding the above fallacies). Heuristically,
this is generally achieved through the production of country reports, each of which presents empirical
research findings regarding – in the present case – children, young people and online technologies.
As a first step of our analysis, EU Kids Online participating teams have contributed to the comparative
analysis by producing country reports. To organise the work, and to aid judgements of similarity and
difference, a template for national reports has been produced, which included concrete research questions
and hypotheses, which were derived from the above model of the research field (see www.eukidsonline.net).
The production of these country reports has drawn on the work completed for Work Package 1, in which
national empirical research studies were identified, coded and entered into an online Data Repository
available at the EU Kids Online website. The country reports are also available on the website.
Type 2: Countries as context for examining general hypotheses
This type of comparative analysis treats each country as a case study with which to test general theoretical
models under different cultural conditions. It focuses on the assumption of similarities across countries, with
cross-national differences thus challenging or limiting pan-national claims. As for type 1, this analysis may be
modest in its attempt to capture the complexity of each country compared, but it is more ambitious insofar as
it seeks to test the hypothesised universality of a particular phenomenon, pooling findings from many
countries in order to establish whether and how an abstract theory applies in each one of those countries.
General hypotheses that were examined concern age trajectories (on the assumption that children develop
into teenagers and adolescents similarly across Europe) or gender differences (again, shown to be fairly
similar cross-nationally by a range of research) or, additionally, parental mediation of online use by children.
Clearly, this type of analysis requires directly comparable data in each country, and the EU Kids Online
network had to consider the extent to which such data are available in all participating countries.
As the second step of our approach, some network members focused on single research questions and
compared the evidence provided by the country reports. In doing so they examined to what extent the results
from different countries support the original hypotheses or provide similar answers to the open research
questions. In the case of differences between countries they discussed possible classifications of countries,
which systematically reflect these differences.
As the third step of our approach some members of the network were nominated to write a summary report
on one of the main sections of the model of the research field (access and usage, risks and opportunities,
attitudes and skills, mediation by parents, teachers and peers). In doing this the authors also checked
whether there is – beyond the evidence from national research – some kind of evidence provided by
international studies, particularly the Special Eurobarometer on Safer Internet issues of 2005/06 (EC, 2006).
These summaries are documented as chapter 2 of this report.
Type 3: Countries as units in a multidimensional analysis
This type of analysis seeks to explain patterns of similarities and differences across countries. It thus
prioritises the identification of measurable dimensions (for which there is available data) on which nations
3
Kohn also identifies a fourth level, the transnational, but this is not applied in the present case.
11
vary, and then examines whether these are related systematically to each other or to a particular measure of
concern (e.g. incidence of online risk to children). Each participating nation thereby serves as one unit or
data source, and must provide measures of both potentially explanatory variables (independent variables)
and variables to be explained (dependent variables).
The strength of this approach is that it seeks to understand the diversity of different national contexts,
achieving this by re-presenting the specificity of each country using a common conceptual language (i.e. in
terms of the interrelations among the multiple dimensions on which each country is compared). It then
develops an explanation for observed differences.4
In our approach we organised the comparative research by combining two steps:
!
Firstly, on the basis of the above described analysis of indicators for children’s and teenagers’ online
behaviour, classifications of countries were identified, which represent differences and similarities
between the countries on the level aggregated individual behaviours, e.g. the incidence of online risk
experienced by children and young people, or the incidence of online opportunities taken up by children
and young people, or the nature and extent of parental activities that mediate children’s online activities.
!
Secondly, the EU Kids Online network identified European similarities and differences in macro-societal
factors like a) the media environment, b) ICT regulation, c) the public discourse on children’s Internet
use and possible risks of the Internet, d) general values and attitudes regarding education, childhood,
and technology and e) the educational system (see above, figure 1.1).
For example, it was asked whether certain political conditions or regulatory policies (e.g. pricing policy,
regulation instruments) lead to more or less risk and opportunities in a country. Or, whether a greater degree
of Internet diffusion results in less or more risk to children when they go online.
The procedure was similar to the comparative analysis on the individual level. Some members of the network
were nominated to write a summary report on single contextual factors and to compare the evidence
provided by the country reports. In doing this the authors also checked available international comparative
statistics which allow a classification of the countries. These summaries are documented in chapter 3 of this
report, together with an analysis of the extent to which the contextual factors allow for an explanation of the
differences and similarities of children’s online behaviours and risk experiences as described in chapter 2.
1.8 The organisation of this report
The structure of the report follows the above logic of our comparative approach. Chapter 2 contains the
comparative analyses on research questions and hypotheses on the individual level.
It ends with conclusions a) with regard to theoretical assumptions on online risks and opportunities (ch.
2.5.1), and b) with regard to meaningful classifications of the European countries concerning children’s
online behaviour (ch. 2.5.2).
Chapter 3 describes relevant contextual factors for children’s online activities and discusses to what extent
they explain the aforementioned differences and commonalities in online behaviour.
Chapter 4 provides a short summary of the findings and conclusions.
The report mainly builds on the country reports from all the EU Kids Online partners. In many cases,
references to concrete studies on the national level are not provided in detail here; however the respective
country reports are available on the EU Kids Online website. Whenever there are substantial quotes or
detailed findings from national studies, reference is made to the respective entry in the EU Kids Online Data
Repository, the searchable online database of research on children’s online use and related risks and
opportunities. These references read as “DR #xyz” for the respective number of the study in the data
repository.
4
‘The Children and Their Changing Media Environment; project (Livingstone & Bovill, 2001) exemplifies this approach,
for it sought to understand how systematic differences in education, wealth, parenting, etc. were associated with
differences across countries in children’s media use, including adoption of new media. Thus it examined the correlations
between national wealth (e.g. GDP), or degree of ICT diffusion, and the dependent variables of children’s media use; this
model expects to find neither similarities nor differences, simply, but rather to find a model that applies across all nations
that explains differences observed among them.
12
2. Children and young people’s internet use:
Comparing across countries
This chapter compares countries with regard to differences and commonalities in children’s and young
people’s internet use. The empirical basis comparison is provided by international and national studies of
online use and online risks on the individual level. Results are presented in four sub-chapters: access and
usage (2.1), risks and opportunities (2.2), attitudes and skills (2.3), and mediation by parents, teachers and
peers (2.4). Each subchapter reviews the general descriptive evidence regarding the key variables across
countries. We then analyse the extent to which these criteria vary according to age, gender, SES, etc.
Subchapter 2.5 will discuss the theoretical consequences of the comparative analysis: Which hypotheses
have been confirmed, which hypotheses need further investigation, which ones have been falsified, and how
can all this be integrated in one coherent model? Secondly this chapter will discuss the issue of classifying
countries on the basis of how children use online media and which kinds of risks they experience. This
classification will provide one part of the input for chapter 3.
2.1 Access and usage5
2.1.1
Parents’ and children’s internet use
Research question R2.1.1: What/how much access to the internet and online technologies do children have?
Access to the Internet and other online technologies is the necessary condition for any use and any risk and
opportunity for children. During the diffusion process from its early stages through to market saturation,
measures of access differentiate among households (and countries) in terms of digital inclusion or exclusion.
As the diffusion process proceeds, concrete aspects of internet access (bandwidth, flat rates etc.) become
more relevant than the simple fact of access. Some information on how many children are able to access the
internet can be inferred on the aggregate level from national statistics on the percentage of households and
schools connected to the internet (see chapters 3.1 and 3.5). Beyond this indirect evidence the search for
comparable data for the countries regarding children’s potential access led to an incomplete and fragmented
picture composed of evidence from national surveys with different instruments, different time frames and
different populations. Since this weak empirical basis does not provide substantial information regarding
differences between the countries the following analysis will focus on the question how many children and
young people actually use the internet and where they use it.
Since the Eurobarometer of 2005/2006 (EC, 2006) provides comparable data for most of the countries
involved in the EU Kids Online project (except Norway and Iceland), these data will be taken as the point of
reference; empirical evidence from national studies will be considered when they hint at shortcomings of the
Eurobarometer data or promise to go beyond the respective comparative findings. The data provided by the
Eurobarometer survey are based on parents’ answers only; as has been discussed in more detail in another
report of the EU Kids Online network (Lobe, Livingstone, & Haddon, 2007), the results of parents’ and
children’s interviews can differ quite substantially, thus the Eurobarometer data have to be interpreted as one
indicator for the respective phenomenon only.
Research question R2.1.2: How much use of the internet and online technologies do children make?
st
In late 2005 the number of children using the internet varied considerably across Europe (see table 2.1, 1
column). While in some countries between two thirds and three quarters of the children went online, the
respective value for other countries was one third or even less. By and large, these differences are reflected
by the empirical evidence provided by other studies on the national level. Thus we take this important
indicator as a criterion to define three country groups 6:
5
This chapter has been written by Uwe Hasebrink based on comparative analyses by Vaclav Stetka (R2.1.1, R2.1.2),
Uwe Hasebrink (H2.1.1, H2.1.2, R2.1.4), Helen McQuillan (H2.1.3), and Cédric Fluckiger & Benoit Lelong (H2.1.4).
6
For this classification we only include the countries, which are involved in the EU Kids Online project.
13
! Group 1 includes those countries involved in the EU Kids Online project, in which more than 65 per cent
of all children use the internet (BE, DK, EE, NL, SE, UK); on average 68 per cent of the children in these
countries use the internet; as evidence from other studies shows, Iceland and Norway also belong to this
group;
! Group 2: countries, in which more than 40 and less than 57 per cent use the internet (AT, CZ, DE, FR, IE,
PL, SI); on average 50 per cent internet users;
! Group 3: countries, in which less than 39 per cent use the Internet (BL, CY, ES, GR, IT, PT); on average
33 per cent internet users.
Table 2.1: Internet use among adults, parents and children
(in order of children’s internet use)
Child’s
use
1)
(%)
Adults’ use of the
2)
internet (%)
Only
parents
EU 25
Netherlands
Denmark
Estonia
Norway**
Iceland*
Finland
Sweden
Belgium
United Kingdom
Luxembourg
Czech Republic
Malta
Slovenia
France
Latvia
Lithuania
Austria
Germany
Poland
Slovak Republic
Hungary
Ireland
Portugal
Spain
Italy
Cyprus
Bulgaria
Romania
Greece
50
72
70
68
68
67
66
66
65
65
60
57
57
57
54
53
53
51
47
47
45
43
42
38
37
36
33
29
28
26
65
97
95
83
100
97
97
98
80
67
83
66
39
71
68
53
48
75
78
42
55
40
59
32
50
55
37
35
34
30
All
respond
-ents
47
85
76
54
80
86
69
85
63
58
57
50
31
53
53
40
32
54
60
34
43
29
51
23
41
49
30
30
31
25
Who uses the internet –
parents and/or child? (%)
Neither
parents
nor child
23
2
3
6
0
3
1
1
10
17
8
21
34
17
17
27
28
15
15
37
27
39
29
49
37
37
43
52
53
54
Only
parents
27
26
27
26
33
30
33
33
25
19
33
22
9
26
29
20
19
33
38
16
28
17
28
13
26
27
25
19
20
20
Only
child
12
2
3
11
0
0
3
1
9
16
9
13
27
13
15
20
24
9
8
21
17
21
12
20
14
8
21
13
14
16
Parents
as well
as child
38
71
68
57
68
67
63
65
55
49
51
44
30
44
39
33
29
42
39
26
28
23
30
19
23
28
12
16
14
10
Source: Eurobarometer 64.4 – Special No. 250: Safer Internet, December 2005; basis: parents/guardians with children
less than 18 years.
*Estimate for Iceland 2005, based on SAFT parent surveys in 2003 and 2007
**Estimate for Norway 2005, based on SAFT parent survey in 2005
1)
All children whose parents claim that their child uses the internet at any place (recoded variable based on QC4.1-6).
2)
All adults who rejected the statement „I do not use the internet“ (QC1.6).
The relative position of the countries regarding children’s likelihood to use the internet is confirmed by
another internationally comparative data base, which has been collected in the framework of a WHO study
(see figure 2.1). Deviations between the two sources can be observed particularly for Portugal and Bulgaria
(the WHO study provides higher percentages than Eurobarometer), and Ireland (the WHO study provides
lower percentages).
14
Figure 2.1: Countries by children’s internet use according to two data sources
% Children aged 16 who spend more than 2 hours using
computers on a normal weekday
80
70
NL
IS
60
SE
GB
PL
50
PT
DE
BG
SK
40
RO
30
ES
AT
HU
LU
LV
FR
FI
BE
EE
NO
DK
SI
CZ
LT
2
R = 0,48
IT
20
IE
GR
10
0
0
10
20
30
40
50
60
70
80
% Children aged 17 and younger who use internet
Sources: a) percentage of children aged 17 and younger using the internet: Eurobarometer 64.4; proportion of children
who spend two hours or more using a computer (Currie et al., 2008).
Research question R2.1.3: What is the relation between parents’ and children’s internet use?
Table 2.1 reveals the relationship between children’s and their parents’ internet use. There is an obvious
correlation between parents’ and children’s internet use across Europe (see figure 2.2).7 In most countries,
the number of parents online is higher than the number of children; on the European level this difference is
15 per cent (parents: 65%, children: 50%). There are some exceptions: in Poland and Portugal (and in
Lithuania, Latvia, and Malta) children are more likely to use the internet than their parents.
7
The Pearson correlation between columns 1 and 2 across the 27 countries is r=.84.
15
Figure 2.2: Countries by parents’ and children’s internet use
80
% Children aged 17 and younger who use internet
NL
DK
70
EE
GB
60
MT
CZ
LT
50
PL
HU
40
30
GR
20
FR
ES
LU
FI
SE
AT
DE
SK
PT
2
R = 0,73
LV
IS
BE
SI
NO
IE
IT
CY
BG
RO
10
0
0
10
20
30
40
50
60
70
80
90
100
% Parents who use internet
Source: Eurobarometer 64.4.
The overall result seems to contradict the expectation that children are more likely to be online users than
their parents’ generation. This contradiction can be solved by looking at the age of the children: The higher
likelihood for parents to use the internet is only based on parents with younger children: For parents with
children less than six years the values are 63% for parents and 9 % for children; for parents with children
between 6 and 11 years the values are 64% and 51% respectively. However, for the oldest group the
opposite relation can be observed: 87 per cent of young people between 12 and 17 years are online and
only 65 per cent of their parents.
Hypothesis H2.1.1: Children whose parents use the internet are more likely to use the internet themselves.
On the European level children whose parents claim to use the Internet themselves are significantly more
likely to use the Internet (58%) than those children whose parents do not use the Internet (34%). This finding
can be confirmed for most of the countries involved in the Eurobarometer study. The only exceptions are
Finland, Sweden, Estonia. In addition the findings from the Netherlands and Denmark are only marginally
significant. This group includes exactly those countries, in which almost all parents use the internet, thus
there is no relevant difference anymore: all children grow up with parents who use the internet anywhere.
Beyond these five countries Cyprus remains a noticeable exception, where the internet use of parents
makes no difference at all. At the other end of the spectrum we find the following countries with substantial
differences in children’s internet use depending on their parents’ internet use: Italy (difference between the
groups: 32%) 8, Portugal (30%), Czech Republic (29%), Poland (26%), Bulgaria (25%). In these countries
children’s internet use depends heavily on their parents’ support.. This might be interpreted as indicator of
lower public support for internet access in these countries.
So far we have compared the general use of the internet, no matter whether parents and children use it at
home or at another place. An even higher correlation can be observed between parents’ and children’s
internet use at home: There are highly significant results for all countries indicating that children whose
parents access the internet at home are more likely to use the internet at home than children whose parents
do not use the internet. On the European level this difference is remarkable: whereas 61% of all children
whose parents use the internet at home use it themselves, the figure for the other group is only 9%.
8
Explanation for the example of Italy: 50.6 per cent of the children whose parents use the internet use the internet
themselves; 18.3 per cent of the children whose parents do not use the internet use the internet themselves.
16
These general findings support the assumption that parents’ behaviour is an important determinant of
children’s internet use. However, the correlation might also reflect the fact that parents follow their children’s
needs and wishes and try to provide what they regard as optimal conditions for their children’s development.
Following the public discourse on the relevance of the internet in today’s society many parents believe that
their children should be familiar with the internet as early as possible. There is some evidence that the
relationship between parents’ and children’s internet use can be interpreted from this perspective. As
research has shown in many countries, households with children are better equipped with media devices;
this is indirectly confirmed by the Eurobarometer data: Among those who have at least one child in their
household, the percentage of internet users in the EU of 25 member states is almost 20 per cent higher
nd
rd
(65%, 2 column of table 2.1) than for the average population (47%, 3 column). This kind of difference can
be observed in all countries.
To get another indicator which can support the hypothesis that parents tend to get access to the internet in
order to provide opportunities for their children, we analysed whether parents’ internet use at home varies
with the child’s age. The rationale is as follows: if parents’ internet use at home was completely determined
by their individual needs there should be no difference between parents of younger and older children.
However, if parents turned to the internet when they think their children should have access or their children
ask to have access, parents of older children should be more likely to use the internet.
The latter thesis is confirmed on the European level: on average 49 per cent of the parents claim to use the
internet at home. This figure is higher for parents with children between 12 and 17 years (52%) and lower for
parents with children younger than six years (44%). Bearing in mind the fact that this analysis cannot
consider the age of the other children in the household, this significant finding supports the assumption that
there is no unidirectional influence from the parents to their children, but also an influence of children on their
parents’ behaviour. However there are also alternative explanations for this finding. Firstly, family income
usually increases as the parents get older and therefore internet access could be an expense postponed by
younger parents. Secondly, parents often have easy access to the internet at work and therefore might feel
no need to connect the home until the child is old enough to want to go online.
To sum up these analyses, the four columns on the right hand side of table 2.1 show the four logical
combinations of parents’ and children’s online use. On the European level the biggest group is made up by
those cases where the parents as well as their children are online (38%). For more than a quarter of the
cases we find the constellation of the parents being online and no internet use of the child. In another group
of less than a quarter neither parents nor children are online. The remaining 12 per cent refers to children
who use the internet whereas their parents do not. Across the countries the distribution of these four groups
necessarily varies with the general percentages of internet users in columns 1 and 2.
What merits further consideration regarding online risks and how children and parents cope with them is the
rd
th
unequal distribution of the 3 and 4 groups in different countries: In some countries, particularly in Sweden,
the Netherlands, and Denmark (and also, as can be inferred from the SAFT survey, Iceland and Norway),
almost all children who use the internet have parents who are also familiar with the online world and thus can
advise or at least understand what their children are doing. At the other side of the spectrum the majority of
children online in Greece, Cyprus and Portugal, i.e. three countries with only a few children online, cannot
count on their parents’ competences since they do not use it. For these countries, the impression is that
there is dissociation between children’s internet use and parental behaviour. This dimension does not simply
reflect the likelihood of children and parents being online in the different countries, because there are some
other countries with a low likelihood of children (and parents) being online, which do not have this high
proportion of children who cannot learn from their parents (e.g. 22% in Italy).
2.1.2
Locations of children’s internet use
These analyses raise the question of where children in Europe have access to and use the internet.
Research question R2.1.4: Where do children in Europe use the internet?
With respect to the question where children in Europe use the internet, the most important locations are their
home (34%, see table 2.2) and school (33%). In addition quite a few children use the internet at a friend’s
home (16%). On the European level the other options are only marginally relevant: at somebody else’s home
(5%), in an internet café (3%), in a library or other public place (4%) or somewhere else (2%).
17
Based on the aforementioned locations for internet use an index has been calculated, which reflects the
st
number of locations where children use the internet (see table 2.2, 1 column). On average young internet
users in the EU go online at 1.9 locations. Again there are noticeable differences between the countries; the
extremes are Sweden (2.8 locations) and Denmark (2.7) on the one side and Italy (1.3), Bulgaria, Greece
and Spain (1.4 respectively) on the other side. Since this index has been calculated on the basis only of
those children who use the internet, the results are not confounded with the general likelihood of internet
use. The strong positive correlation across the 27 countries listed in table 2.1 between internet use and the
number of locations where children go online (r=.88) reflects the general diffusion process of the internet. In
those countries where the diffusion process started earlier, the higher number of children who are online
goes along with a more diverse set of locations where children use the internet.
Looking at single countries we find interesting differences where children use the internet. One striking
nd
rd
difference can be observed (see the shaded cells in table 2.2, 2 and 3 column, and figure 2.3) between
countries where the internet is more often used at home (the Netherlands, Denmark, Sweden, Belgium,
Slovenia, Germany, Spain and Italy) and others where the school is the most important place for children to
go online (the UK, the Czech Republic, Poland, Portugal, Bulgaria, Romania, and Greece); in Estonia,
France, Austria, and Ireland both places are equally important.
Figure 2.3: Countries by location of children’s internet use
80
% Children who use internet at school
70
60
NO NL
GB
FI
50
CZ
40
SK
30
HU
2
R = 0,68
10
AT
IS
SE
EE
BE
SI
LV
IE
PT
20
BG
RO
LT
PL
MT
DK
LU
FR
DE
CY
ES
GR
IT
0
0
10
20
30
40
50
60
70
80
% Children who use internet at home
Source: Eurobarometer 64.4.
Further relevant differences are related to internet cafés on the one hand and libraries or other public places
on the other hand. Internet cafés, i.e. the most commercialized option to go online, are most often used in
countries with generally low percentages of internet users and particularly low figures for libraries and other
public places (e.g. Bulgaria and Poland). On the other side, the Nordic countries obviously provide better
online access at public places.
In order to provide a simple indicator for the comparison across the countries involved, we defined two
groups of internet users: a) those who access the internet at home and, maybe, also at other places; b)
those who access the internet at school and, maybe, at other places, but not at home. On the European level
68 per cent of all young internet users belong to the first group, they have access to the internet at home and
at other places; 32 per cent do not use the internet at home but at school or any other place. Looking at the
results on the country level leads to a classification of three groups:9
9
Again, this classification does only include the countries, which are involved in the EU Kids Online project.
18
!
Group 1: Countries, in which more than 80 per cent of the young online users access the internet at
home. In this respect the highest values are reached in Norway (100%), Iceland (99%), the Netherlands
(96%), Sweden (93%), and Denmark (91%), other members of this group are Slovenia (83%), Germany
(83%), and Belgium (81%).
!
Group 2: Countries with more than 58 and less than 70 per cent using the internet at home. The countries
in this group are the UK and Estonia with 60 per cent young internet users who are online at home, the
following countries are Ireland, Austria, France, Cyprus, Czech Republic, Italy, and Spain (59%).
!
Group 3: Countries with less than 50 per cent of young online users who use the internet at home. These
are Poland (48%), Greece (44%), Portugal (42%), and Bulgaria (27%).
Table 2.2: Children’s online use at different places
(in order of children’s internet use at any place)
Average
number of
locations
EU 25
Netherlands
Denmark
Estonia
Norway**
Iceland*
Finland
Sweden
Belgium
United Kingdom
Luxembourg
Czech Republic
Malta
Slovenia
France
Latvia
Lithuania
Austria
Germany
Poland
Slovak Republic
Hungary
Ireland
Portugal
Spain
Italy
Cyprus
Bulgaria
Romania
Greece
1.9
2.7
2.7
2.2
nd
nd
2.4
2.8
2.1
2.2
2.0
1.9
1.9
2.0
1.9
1.8
1.7
2.0
1.9
1.6
1.8
1.6
1.7
1.5
1.4
1.3
1.7
1.4
1.6
1.4
At
home*
**)
At
school
***)
At a
friend
’s
home
34
69
64
47
67
66
57
61
52
45
54
35
40
47
35
28
22
35
39
22
14
17
28
16
22
22
20
8
8
12
33
57
53
45
55
54
46
53
36
58
35
45
45
35
33
34
35
35
26
33
39
32
28
27
16
15
19
12
13
15
16
43
36
30
9
11
30
45
26
20
20
17
12
17
23
19
18
22
16
9
12
9
9
6
4
5
11
4
6
4
At
somebody
else’s
home
5
17
16
10
nd
nd
9
7
10
11
3
2
4
5
6
4
3
4
1
2
2
3
2
2
1
2
2
2
5
1
In an
internet
café
3
1
3
5
nd
nd
1
4
3
2
3
2
3
0
1
5
8
4
2
8
8
1
0
1
3
1
1
15
11
4
In a
library or
other
public
place
4
5
11
11
1
4
14
10
6
9
4
6
2
8
4
8
5
2
1
3
4
4
3
4
1
2
0
0
1
0
Somewhere
else
2
2
5
3
Nd
Nd
2
4
1
2
1
1
0
1
2
1
2
0
2
1
1
1
2
1
3
1
1
0
2
0
Index
“not at
home”
32
4
9
31
0
1
13
7
19
31
9
39
29
17
35
48
59
33
19
53
70
61
33
58
41
39
38
73
71
56
less than 18 years.
* Estimate for Iceland 2005, based on SAFT parent surveys in 2003 and 2007
** Estimate for Norway 2005, based on SAFT parent survey in 2005
*** Shaded cells in column 2 and 3 indicate that the value is higher (at least 3%) than the value in the other column.
These groups are quite similar to the classification on the basis of general internet use,10 which indicates that
there is a general diffusion process of the internet, with internet access at home becoming the normal
10
The Pearson correlation across the 27 countries between the percentage of children being online and the percentage
of “not accessing the internet at home” is r=-.76.
19
condition for children and young people. However, there are some remarkable differences between the
classifications. Within the group with the highest proportion of internet users, Estonia and the UK are less
focused on home use than the other countries; this might be an indicator for a particularly ambitious public
policy, which supports internet use at schools and other public places.
Within the second group, Slovenia is much more home oriented than the other countries with similar
percentages of online users (Austria, France and the Czech Republic); and for the rest of this group there
are completely heterogeneous results regarding the relevance of home and school use: Whereas in
Germany only 19 per cent of the internet users are not online at home, this figure is much higher in Ireland
(33%) and particularly in Poland (53%).
Finally, the third group is divided into Cyprus, Italy and Spain with a moderate relation between home and
school use on the one hand, and Bulgaria, Portugal and Greece with very high percentages of young internet
users who do not use the internet at home but at school or elsewhere. To what extent these differences
between the countries can be explained by different political and cultural conditions will be examined in
chapter 3.
In all, the European countries differ quite substantially regarding the place where children use the internet.
Since it is likely that the places where children are online are connected with specific risks, the countries
provide quite different conditions for potential harmful experiences and for political and pedagogical means
to support a safer use of the internet; this will be discussed in more details in the next chapters.
2.1.3
Positional factors influencing children’s internet use
The following analyses focus on the question of the extent to which basic positional variables shape access
to online media and the frequency and duration of use. The factors to be analysed here are age, gender, and
socio-economic status.
Hypothesis H2.1.2: As children get older their access to and use of the Internet and online technologies
increases.
The evidence as provided by the national reports refers to different empirical levels:
!
Children’s access to the Internet,
!
Time and frequency of use of the Internet in general,
!
Time and frequency of use of specific online applications,
!
Time and frequency of use of the Internet at different places.
In general all evidence provided for all countries (if available) supports the hypothesis that there is a steady
increase of access to and use of the internet and online technologies with age. Beyond this general result,
for some of the countries there are some contradicting or contextualizing findings:
!
Findings from countries with higher internet penetration indicate that the increase of access to and
frequency of use of online media stops around the age of 12/13 years; this is due to a ceiling effect,
since from this age on almost all young people use online media.
!
For those countries which report contradicting or ambivalent findings, there are specific reasons. In the
Czech Republic the fact that survey respondents were 12+ years, might have resulted in the absence of
no age differences (see the argument above). In Poland there is evidence that 13-15 years old young
people use the Internet more often than those aged 16-18. A similar finding is provided for the UK:
those aged 16/17 years old use the internet more often than the older age group.
!
The French report hypothesizes that the difference between 9-10 and 12-14 old children regarding the
amount of use (and not the frequency of use) is due to parental mediation: for the younger group the
parents restrict the time their children spend with the Internet.
!
!
The Greek report emphasizes that the range of services used clearly increases with age.
According to a hypothesis from the Polish report, one reason for a decrease of online use between
13/14 and older is specifically the decreasing interest in (online) gaming.
20
The comparative analysis runs into problems in so far as the respective findings refer to different age bands
and different indicators of use. Thus it is difficult to analyse the development on a more detailed level. For
this reason, a secondary analysis of Eurobarometer data has been run, in order to have some indicators on
the basis of comparable data and a large sample, which allows for the analysis of small age bands. The
limitations of these data are twofold: they are based on statements of the parents only, and they refer to the
simple fact, whether children have used the internet, whereas there is no indicator for the frequency and
amount of use.
According to these data the proportion of children who use the internet increases significantly as they grow
older (see table 2.3). It is interesting to note that the peak of the diffusion of the Internet is already reached at
about 12 or 13 years, after that there is only a very small increase. The figures for boys and girls are quite
similar. The small overall difference is based on children between 6 and 9 years: boys seem to start using
the internet earlier, whereas there are no difference between 10 to 13 years old boys and girls.
Table 2.3: Internet use by age and gender
(EU 25, %)
< 6 years
6-7 years
8-9 years
10-11 years
12-13 years
14-15 years
16-17 years
Total
9
34
51
68
85
87
88
Boys
9
37
55
67
85
90
88
Girls
9
30
47
69
84
84
88
Source: Eurobarometer 64.4 – Special No. 250:
Safer Internet, December 2005;
basis: parents/guardians with children less than 18 years.
Since these aggregate figures might hide specific developments dependent on the general internet diffusion
in the countries, the respective figures have been calculated for the three groups of countries, which have
been defined above on the basis of the percentage of children who are online.
Figure 2.4 shows that all three groups show similar figures in two respects. Firstly, the use of the Internet
steadily increases with age. Secondly, after the age of 12-13 years this increase is only very small – if at all.
Figure 2.4: Children who have used the Internet by age and country groups
(in per cent of all children)
100
90
80
70
60
50
40
EU 25
Group 1
Group 2
Group 3
30
20
10
0
<= 5 yrs.
6-7 yrs.
8-9 yrs.
10-11 yrs.
12-13 yrs.
14-15 yrs.
16-17 yrs.
less than 18 years.
21
Beyond these similarities, there is a significant difference between the three country groups: in the countries
with a higher internet use children are much younger when they start to use the internet. As a consequence
one can say that online users in countries with a high level of internet use are younger (for group 1 the
average age is 11.5 years) than online users in countries with a lower level of internet use (for group 4 the
average age is 12.7 years).
As these data show, the general hypothesis is widely supported. On the European level, there is a steady
increase in the proportion of children who use the Internet; the older the children the more likely they are to
use the internet. Beyond that there is evidence that the increase with age reaches a plateau quite early: the
differences between the group of 12-/13 years old children and the older groups are rather small – if at all.
However, when looking at country groups, which differ regarding their general internet penetration, quite
specific patterns of age related developments can be observed. The difference between countries with a
high versus a low amount of internet use is mainly based on the younger age groups. In countries with a
lower level of internet use, children start to get access to and to use the internet later. As a consequence,
online users in these countries are older than online users in countries with a high level of online use.
Regarding the promotion of safer internet use, the fact that the internet is already a normal tool for children at
the age of ten years and increasingly becoming an attractive tool for those between 6 and 10 years
emphasizes the need to develop measures supporting Safer Internet for all age groups – according to the
respective functions, for which children go online. Until now, only few younger children have used online
media in those countries with lower internet penetration; in these countries the target group is older.
Dependent on further findings on how age differences affect which risks children encounter (see chapter
2.2), this difference in age must be considered when promoting Safer Internet in different countries.
Hypothesis H2.1.3: There are no gender differences in children’s access to or amount of use of online
technologies.
The hypothesis that there are (no longer) gender differences regarding children’s access to and use of the
internet is fully supported in only one country: Denmark. Evidence in 11 countries contradicts the hypothesis:
Poland, Norway, Austria, Belgium, Bulgaria, Czech Republic, Sweden, France, Portugal, Slovenia, and the
Netherlands. Contradictory results come from 6 countries, with evidence that both supports and challenges
the hypothesis: Estonia, Germany, Iceland, Ireland, Italy and the UK. No evidence is available for three
countries: Greece, Spain and Cyprus.
According to the reports gender gaps in access to the internet are small and are closing in nearly all
countries, particularly for younger children. In the UK, this pattern is more evident among older children.
However, in almost every country, boys are more likely than girls to spend greater amounts of time online,
have more places to access the internet from, have their own computer and internet access and have
access to a PC and internet in their bedrooms.
Several patterns are emerging:
!
Gender gaps in access are diminishing as home and school internet access becomes more
commonplace.
!
There is a growing bedroom culture for teenagers and solitary use of the internet is increasing,
particularly for boys.
!
The amount of time spent by boys and girls online is increasing in all countries.
Age seems to be crucial. Consideration needs to be given to the examination of different age groups.
Further evidence on this issue can be drawn from the Eurobarometer survey. Table 2.4 shows how many
girls and boys in different age groups use the internet at which location. There is a slightly higher percentage
of boys (52%) who use the internet at any place (compared to 49% girls). This difference comes out to be
significant for internet use at home and at a friend’s place. As seen above the small difference is mainly due
to clear differences in the age group of 6 to 9 years, when boys are more likely to use the internet.
22
Regarding the age groups the table shows some age specific patterns. Due to the very rare internet use of
children younger than six years, there are no gender differences in this age group. For those between six
and eleven years significantly higher percentages are found for boys being online at home and at school. In
the oldest group boys are yet more likely to be online at home; in addition more of them use the internet at a
friend’s place. However girls are more likely to use the internet at libraries or other public places.
Table 2.4: Internet use of girls vs. boys at different places by age
11
(in %, basis: EU 25; shaded cells indicate significant differences (p<.05) )
Age
0-5 years
6-11 years
12-17 years
All children
At any
place
9:9
48 : 53
85 : 88
49 : 52
At home
7:7
30 : 34
59 : 63
33 : 36
At school
3:3
31 : 34
62 : 61
32 : 34
At a
friend’s
place
1:1
12 : 13
31 : 35
15 : 17
At an
internet
café
0:0
1:1
6:8
2:3
At a library
0:0
3:2
9:7
4:3
less than 18 years.
Looking at the results for the single countries leads to the conclusion that in most of the countries the
percentage of boys being online is slightly higher than the percentage of girls. However, in most cases these
differences are not statistically significant; the only exceptions here are Austria (boys: 62%, girls: 45%), Italy
(boys: 41%, girls: 30), and Germany (boys: 52%, girls: 43%). As evidence from national studies, e.g. from
Germany, shows, these differences are likely to decrease with the ongoing diffusion process.
The only countries where, according to the Eurobarometer survey, internet use is slightly (not statistically
significant) more common for girls than for boys are Ireland, Netherlands, Portugal and the United Kingdom.
To test the hypothesis that gender differences are greater in the early stages of internet diffusion, the
percentages of boys and girls using the internet were calculated for the three groups of countries (defined
above on the basis of the percentage of internet users; table 2.5). The results indicate that the overall
difference between boys and girls disappears in the later stages of the diffusion process, since a gender gap
can be observed only in groups 2 and 3 where the diffusion is less advanced than in group 1.
Table 2.5: Gender differences in internet use in countries with high, medium and low internet use
Boys
Group 1 (“High internet use”)
Group 2 (“Medium internet use”)
Group 3 (“Low internet use”)
Total EU 25
68.6
52.0
35.0
52.0
Girls
68.0
48.2
31.3
49.0
less than 18 years.
Another indicator to assess gender differences is the variety of locations where children have access to the
internet (see above). The analysis was run on the basis of only those children who actually use the internet,
thus the following results do not repeat country differences in the number of children who are online, but just
reflect the number of different locations of access. On the European level we find exactly the same values for
girls and boys: On average they are online at 1.9 different places. In a multivariate analysis with gender and
age as independent variables and the variety of locations as dependent variable there is no main effect of
gender and no interaction between gender and age.
Hypothesis H2.1.4: There are inequalities in access as a consequence of inequalities in SES (socioeconomic
status e.g. household income, parental education, social class).
Since access issues are addressed by many quantitative surveys in many countries, there is quite a lot of
empirical evidence regarding SES and inequalities in children’s access to the internet. In almost all countries,
there is evidence to support the hypothesis of a correlation between access and SES. The inequalities in
11
It has to be noted that – due to the high number of respondents (n=7,560) – even quite small differences between the
groups qualify as statistically significant.
23
SES can be measured by or assumed on the basis of the household income, the parents’ educational level,
or the type of school where children go. The exact figures vary between the countries, however, in most
countries there is a significant difference between higher and lower class children. In countries where the
internet and computers are widespread, the ownership of a computer can reach 80 to 90% for higher class
children, compared to 50 to 60% for working class children.
Only two countries report evidence that contradicts the hypothesis. This is the case for Iceland, due to the
fact that “the Icelandic society does not seem to follow the same patterns of differentiation as most European
countries”, and Sweden, where no difference in internet access between social groups could be found,
though “more research is needed”, as it is not clear who the 10% of children without a computer are.
As the country reports regarding social inequalities show, future research should more systematically
differentiate the various dimensions of SES: household income, level of education and profession of parents,
etc. Migrants and ethnic minorities should be included in the survey samples, in the data analysis as well as
in the final reports. Each family has different resources (social capital, economic capital, cultural capital)
whose specific role could thus be more easily analyzed.
As exemplified in the UK report, research could provide a more accurate view on the access issue, e.g. not
only whether children access the Internet or not, or the frequency of access, but also whether it is at home
(vs. at school or in other places such as neighbours' or friends' home), in the child's bedroom, in a room
shared (or not) with other family members. Research should also distinguish broadband and dial-up access,
since this difference has proven to have a great impact in terms of shaping uses and the learning process.
Such an improved definition of access might explain the apparent absence of evidence in Sweden. In
countries with a high diffusion rate, the role of SES may be more fruitfully searched in the quality (rather than
the mere existence) of internet access.
Access to online services from a mobile phone is seldom taken into account in the data provided. More
research should be done on that topic, because the diffusion of mobile access has already started. New risks
and opportunities will probably appear with these individual (and easy to carry) devices.
For some national reports (Italy for example), the actual use of internet is a part of the access issue. This
illustrates the partial overlap between hypotheses H2.1.4 and H2.1.5. The issue of children living in a home
with Internet access but not using it should indeed be addressed.
Hypothesis H2.1.5: There are inequalities in online use as a consequence of inequalities in SES.
Though evidence could not be found in a few countries, it seems that there is a general agreement
throughout European countries that there is a correlation between SES and the frequency and amount of
online use. In some countries (Estonia, France, Sweden), there is no significant difference in the frequency
of use. On the other hand, Iceland, Norway and the UK report that children of parents with a higher
education and/or belonging to a higher class do use the computer more often than the other children.
Some of the often contradictory findings regarding the influence of SES on children’s frequency of internet
use may arise from the fact that no difference is made between different places of use and between different
types of use; this will discussed in more detail in chapter 2.2.
2.2. Risks and opportunities12
The many hopes and fears regarding the opportunities that the internet can offer to children and young
people, along with its attendant risks, have attracted considerable attention across Europe and elsewhere.
The result is a series of pressing questions for policy makers, regulators, industry and the public about
whether, in practice, young people are taking up these opportunities, whether some are benefiting more than
others, and which factors might facilitate the beneficial uses of the internet in an equitable manner. These
opportunities are widely judged to include entertainment, information, education, communication, networking,
12
This chapter has been written by Sonia Livingstone based on comparative analyses done by Pille PruulmannVengerfeldt (R2.2.1), Yiannis Laouris (H2.2.1), Bojana Lobe (H2.2.2), Helen McQuillan (H2.2.3, H2.2.4), Cédric Fluckiger
& Benoit Lelong (H2.2.3, H2.2.5), Uwe Hasebrink (H2.2.6), and Sonia Livingstone (R2.2.2, R2.2.4, R2.2.5).
24
creativity, play and civic participation – a heterogeneous set of activities for which there is considerable
optimism and public/private sector provision.
Equally pressing, however, are the questions regarding whether young people are encountering risks online,
whether some are particularly at risk, and which factors might mitigate against the risks of internet use.
These risks, also encompassing a heterogeneous set of intended and unintended experiences, include
encountering pornographic, self-harm, violent, racist or hateful contents online, inappropriate or potentially
harmful contact via grooming or harassment, and, attracting recent attention, problematic conduct among
peers such as bullying, ‘happy slapping’ or privacy invasions of one kind or another.
Widely understood as mutual opposites, there is nonetheless considerable scope for interpretation and
contestation – both conceptually and between adults and children – regarding the classification of specific
activities as either opportunities or risks.
Public policy regarding children and the internet is framed by the coincidence of three factors: first, the
extraordinary rapidity of the internet’s diffusion and development, faster than any previous media and so
outpacing society’s ability to adjust; second, an endemic cultural fear of the new, encouraged by media
panics framing the internet as impossible to regulate as a source of threats to children’s safety; and third, the
novelty of a reverse generation gap whereby parental (and teachers’) expertise is exceeded by children’s
ability both to use the technology and evade adult management.
This chapter undertakes the following:
!
!
!
!
!
!
An examination of evidence from across Europe relevant to two main research questions – what are the
main opportunities, and the main risks, experienced by children online?
To organise these findings, it puts forward a classification of varieties of online opportunities and
varieties of online risks.
Based on the risk findings, the chapter then proposes a classification of countries according to the
incidence of online risk experienced by children.
Following up on the findings for risks, these are examined by children’s age, gender and sociodemographic background, according to hypotheses generated from the published literature.
Available research on how children respond to risk is then examined, including the effect of internet
literacy or skills, and different strategies of coping with risk.
Finally, we consider the relationship between online opportunities and risks.
2.2.1
Opportunities experienced by children online
Research question R2.2.1: What are the main opportunities experienced by children online?
Among the 21 countries included in EU Kids Online, evidence was available from almost all about the main
opportunities experienced by children; however, little evidence was available from Slovenia, Bulgaria and
Greece. In some countries only, evidence was available regarding both adults and children’s perception of
online opportunities – Sweden, Poland, Ireland, Denmark, Greece, Italy, UK and Norway.
In general, adults and children agreed that children use the internet as an educational resource, for
entertainment, games and fun, for searching for global information and for social networking, sharing
experiences with distant others. Other opportunities, such as user-generated content creation or concrete
forms of civic participation, are less common.
!
In the majority of countries (the UK, Sweden, Spain, Poland, Norway, Italy, Ireland, Iceland, Germany,
Estonia, Denmark, Czech Republic, Cyprus and Austria) children perceive entertainment, games and
fun as major opportunities of the internet. In most countries too, children use the internet as an
educational resource (UK, Portugal, Poland, Norway, Italy, Ireland, Iceland, Germany, Estonia,
Denmark, Belgium, Austria,).
!
There is evidence that social networking and sharing experiences with distant others is common among
children from the UK, the Netherlands, Sweden, Spain, Poland, Norway, Italy, Ireland, Iceland,
Germany, France, Estonia, Denmark, Czech Republic, Belgium and Austria.
25
!
The opportunity to search for global information is mentioned by the youth of the UK, Sweden, Spain,
Ireland, France, Germany, Cyprus, Belgium and to a lesser extent also in Estonia. Furthermore, usergenerated content creation is more common in Ireland, Iceland, France and in Belgium and less so
among children in Estonia and the Czech Republic.
!
Parents are more likely to stress online opportunities to access global information (Sweden, Poland,
Italy, Greece) and the use of the Internet as an educational resource (Denmark, Greece, Ireland,
Sweden, Norway, Italy). They tend to underestimate the value to their children of the Internet for social
relationships and entertainment.
Little clear or cross-nationally comparable information was available regarding the incidence and take-up of
these various opportunities, however. Focusing on particular activities or applications more than the
opportunities these may afford, the Mediappro project (Mediaprro, 2006) produced the only directly
comparable data available on children’s uses. This survey of 7393 12-18 year olds regarding their
appropriation of new media in nine European countries found some, not easily interpretable, cross-national
variation in online activities (see table 2.6).
Table 2.6: Children’s online activities
Activities on the Internet (% sometimes/often/very often)
Search
engines
Email
Belgium
95
74
Instant
Messenger
81
Chat rooms
Downloading
Denmark
92
66
87
26
50
Estonia
90
69
88
33
73
France
94
97
69
32
49
Greece
81
46
39
41
65
Italy
86
59
49
33
59
Poland
91
62
75
34
67
Portugal
95
69
77
38
60
UK
98
81
78
20
60
Average
91
66
71
32
60
28
58
Source: Mediappro (Mediappro, 2006) p.12, see www.mediappro.org.
Overall, these figures suggest a fairly constant and familiar picture, with children mixing educational,
entertaining, informational and networking activities in substantial numbers, while tailoring internet use to suit
their interests.
Generally, once they gain access (and skills), it can be concluded that children in all countries prioritise
online communication, various forms of entertainment and play, and information provision, while for parents
the benefits of educational resources are higher on their agenda. There is insufficient evidence, however, to
justify a classification of countries in terms of online opportunities engaged in by children.
If online opportunities are to be increased across Europe, much depends on, first, the child’s role (their
motivation and resources) and the online provision available to them (and, thus, the providers’ motives or
social goals).
Hence we propose a classification of online opportunities for children and young people as follows. In the
table below, the cells shaded grey are those for which a fair body of empirical evidence is already available.
For many other opportunities discussed in public and policy circles, too little is known regarding either
provision or take-up by children as yet.
26
Figure 2.5: Classification of online opportunities
Online
opportunities
Providers’ motives
Child’s role
Education and
learning
Participation and
civic engagement
Creativity
Identity and social
connection
Use of educational
resources (incl.
edutainment)
Use of global
information
Use of diversity of
resources for
creativity and play
Advice (personal /
health / sexual etc)
Contact with others
who share one’s
interests
Exchange among
interest groups,
Being invited or
inspired to
participate in
creative processes
Social networking,
sharing experiences
with distant others
Self-initiated or
collaborative forms
of learning
Concrete forms of
civic engagement
User-generated
content creation
Expression of identity
Content
Child as
recipient
Contact
Child as
participant
Conduct
Child as actor
Livingstone and Helsper propose a graduated sequence of activities towards digital inclusion (Livingstone &
Helsper, 2007). Based on findings for UK 9-19 year olds, differences among users fell into four orderly steps,
suggesting a ladder of online opportunities as follows.
!
Step 1 centres on information-seeking. This is the first step for everyone, and characterises internet use
among those who just take up a few online opportunities. They may be termed basic users.
!
Step 2 adds in games and email. Thus, those who take up a few more opportunities are likely to use the
internet for information, entertainment and communication. These may be termed moderate users.
!
Step 3 adds in instant messaging and downloading music. Those who take up a fair range of
opportunities continue to seek information but they expand their peer-to-peer engagement. They may
be termed broad users.
!
Step 4 adds in a wide range of interactive and creative uses, while continuing the foregoing uses,
making for a diversity of uses among those who take up the most opportunities online. These are
termed all-rounders.
To some degree, children progress ‘up the ladder’ as they get older – most activities online become more
common with age. However, the active promotion of further online opportunities in Europe should, one may
conclude, encourage these steps in turn. For example, providing positive information resources of interest to
children is the best way to encourage beginners. Contrary to the views of many adults, following this up with
the provision of fun games is a good next step. Looking at these steps the other way around, one may
suggest that children not yet comfortable with peer-to-peer communication are unlikely to engage in civic
participation.
In terms of future research priorities, it seems that a systematic approach to data collection is needed if the
take up of opportunities (as judged by children and parents) is to be encouraged in a comparable manner
across Europe. While research has provided a portrait of the activities especially enjoyed by children, it is
less clear about parental views, which matter – and should be further researched – because parents’ beliefs
regarding the benefits of internet use for their children will motivate their provision of hardware and software
resources, their social and technical infrastructural support, their efforts to overcome digital inequalities and
their perception of the likely costs if safety concerns lead them to restrict children’s access.
27
2.2.2
Risks experienced by children online
Research question R2.2.2: What are the main risks experienced by children online?
Developing the three C’s approach, EU Kids Online has classified the array of risks to children as shown in
the table below.
!
The vertical dimension recognises that risks to children derive from the three modes of communication
afforded by the internet: one-to-many (child as recipient of mass distributed content); adult-to child (child
as participant in an interactive situation predominantly driven by adults); and peer-to-peer (child as actor
in an interaction in which s/he may be initiator or perpetrator).
!
The horizontal dimension acknowledges four main forms of risk to children’s development and wellbeing - commercial, aggressive, sexual and value threats.
!
Note that while the specific risks that fall into each cell may change over time, the categories are more
enduring.
Figure 2.6: Classification of online risks
Online risks
Providers’ motives
Child’s role
Commercial
Aggressive
Sexual
Values
Content
Advertising,
spam,
sponsorship
Violent/ hateful
content
Pornographic or
unwelcome sexual
content
Racism, biased or
misleading info/ advice
(e.g. drugs)
Tracking/
harvesting
personal
information
Being bullied,
harassed or
stalked
Received unwanted
sexual comments,
being groomed,
meeting strangers
Self-harm, unwelcome
persuasion
Illegal downloads,
hacking, gambling
Bullying or
harassing
another
Sending or posting
porn, sexual
harassment
Providing advice e.g.
suicide/pro-anorexic
chat
Child as
recipient
Contact
Child as
participant
Conduct
Child as actor
Many potential online risks have been discussed in public, policy and academic circles, but not all have been
researched as yet. Evidence of the incidence, distribution and possible consequences of these types of risk,
on a reliable cross-national basis, is sparse. Risks are particularly difficult to define in culturally-consensual
ways, and they are difficult to research in methodologically-rigorous and ethically-responsible ways. Few
studies are conducted comparatively, and exact samples, methods and measures vary considerably
(Livingstone & Haddon, 2008; Lobe et al., 2007).
Noting strong caveats regarding difficulties with and differences in definitions and methods, the following
portrait of available evidence is offered. It is based on risks as reported by children – generally teenagers
who use the internet (unless other age groups are specified). In high internet access countries, the figures
therefore apply to most of the population. In low internet access countries, however, figures obtained from
online teenagers apply to a smaller, often more urban and/or wealthier, segment of the population.
For the five cells shaded above, sufficient data exists to scope the incidence of online risk – see Table 2.7.
28
Table 2.7: Summary of national reports on evidence for forms of risk
Form of risk encountered by children
and median response across countries
researched
Aggressive content (child as recipient):
…. Seen violent or hateful content
The approximately median response is 32%
of online teenagers who have encountered
this risk in Europe
Incidence by country
Note: Percentages refer to online teenagers unless
otherwise stated
• 90% in Ireland (10-20 yrs – SNS users)
• 51% in Poland (12-17 yrs)
• Up to 40% in Belgium (9-12 yrs)
• 39% in The Netherlands (13-18 yrs)
• 35% in Denmark (9-16)
• Up to 33% in France (12-17 yrs)
• 31% in UK (9-19 yrs)
• 35% in Iceland (9-16 yrs)
• 29% in Norway (9-16 yrs)
• 29% in Germany (on mobile) (12-19 yrs)
• 26% in Sweden (9-16 yrs)
• Up to 25% in Italy (7-11 yrs)
• 15% in Austria (10-15 yrs)
Aggressive contact (child as participant):
…. Been bullied/ harassed/ stalked
The approximately median response is 1520% of online teenagers who have
encountered this risk in Europe
•
•
•
•
•
•
•
•
•
52% in Poland
31% in Estonia (6-14 yrs)
21% (7-11 yrs) and 18% (12-19 yrs) in Italy
20% in UK (11-19 yrs)
19% in Ireland (of chatters 9-16 yrs)
16% in Norway
15% in Iceland (9-16 yrs)
16% in Sweden (9-16 yrs)
10% in Belgium
Aggressive conduct (child as actor):
… Sent bullying/ harassing messages
this risk in Europe
Sexual content (child as recipient):
… Seen pornographic or unwelcome
sexual content
this risk in Europe
•
•
•
•
18% in Belgium
14% in Norway
10% in Denmark
8% in Ireland
•
•
•
•
•
•
•
•
•
•
•
•
80% in Poland
57% in UK (9-19 yrs)
50% in Austria (10-15 yrs – 60% of 11-18 yrs)
47% in Norway (9-16 yrs)
46% in Netherlands (13-18 yrs)
Up to 40% in Belgium (9-12 yrs)
37% in Ireland (9-16 yrs)
Up to 33% in France (12-17 yrs)
29% in Denmark (9-16 yrs)
Up to 25% in Italy (7-11 yrs)
Sexual contact (child as participant):
(1)
Received
unwanted
sexual
comments
this risk in Europe
•
•
•
•
•
•
•
•
56% (of 12-17 yrs meeting strangers online) in Poland
32% in Sweden (9-16 yrs) have received unwanted sexual
comments, and 15% (13-16 yrs) were subject to unwanted
talk about sex
31% in UK (9-19)
25% in Iceland
24% in Norway
9% in Germany (6-13 yrs)
9% in Ireland
6% in Portugal (8-18 yrs)
(2) Met online contact (stranger) offline
•
•
20% in Czech Republic (12-17 yrs)
29
this risk in Europe
•
•
•
•
•
•
•
•
Additionally, a risk associated with most
contact
risks:
… Has given out personal information
this risk in Europe
•
•
•
•
•
•
•
•
•
•
19% (of 12-17 yrs) in Poland
16% in Spain
14% in Norway
9% in Denmark (18% of online chatters, where 48% of
those online are chatters)
8% in UK (9-19 yrs)
8% in Belgium
6% in Estonia (6-14 yrs)
91% in Czech Republic
64% in Poland
16% in Spain
14% in Norway
9% in Denmark (18% of online chatters, where 48% of
those online are chatters)
46% in UK (9-19 yrs)
44% in France (12-17 yrs)
13% in Belgium (9-12 yrs)
Finally, a miscellany of further risks are noted by the national reports, as follows:
!
Viruses/scams/spam – seen as problem by many children
!
17% in Belgium (9-12 yrs) felt threatened online, and 40% felt shocked by online content
!
19% in Estonia (6-14 yrs) disturbed by stranger online
!
40% in Austria (10-15 yrs) visited gambling sites and 11% 11-18 yrs visited suicide forum
!
44% girls/30% boys in Germany (12-19 yrs) had unpleasant experiences in chat rooms, and 29% had
seen a filmed beating
!
19% in Denmark have been harassed/bothered/upset, and 77% of chatters have been insulted
!
15% in Iceland asked for a picture of self naked online
!
16% in Iceland received emails/messages which made them worried or frightened
!
7% in Ireland (10-14 yrs) made uncomfortable by material
!
16% in Italy (13-17 yrs) had unpleasant or bad experiences, 8.1% (12-19 yrs) received threatening
content and 21.7% (12-19 yrs) received fake information about themselves.
!
46% in The Netherlands (13-18 yrs) disturbed by annoying comments when chatting/using IM
!
30-40% of young people in Slovenia are bothered by spam/viruses/slowness of the websites loading;
44% express their concern about Internet safety.
!
11% in Spain have felt afraid online (and 36% tend to disconnect because they are concerned about
other people online)
!
16% in UK seen something frightening/worrying
It seems that online risk attracts public concern and policy attention with justification. In most countries,
significant minorities and, in some cases, a majority of children, especially teenagers, are encountering a
range of aggressive and sexual risks. These include content, contact and conduct risks.
Yet, many pressing questions remain, with little data available in some countries and, as noted earlier, many
difficulties of measurement and comparability impeding a clear picture
Unlike for online opportunities, however, the above table does provide the basis for an approximate
classification of countries in terms of incidence of online risk experienced by children and young people.
30
Although stronger data would be greatly preferable before advancing such a classification, we attempt one
here nonetheless in order to gain some benefit from the available research findings.
To construct this classification, for those countries where several quantitative estimates of risk are available,
the following ‘rough and ready’ calculation was applied. A rank of 1 was given if the national percentage is
below the median for the risk category, a rank of 2 is given if the percentage is close to the median, and a
rank of 3 is given if the percentage is above the median.
!
High risk countries include those new to the internet (Poland, Czech Republic) or experienced with the
internet (Netherlands).
!
Next riskiest are Estonia, Norway, UK.
!
Around the European average for online risk to children – Austria, Belgium, Denmark, Ireland, Sweden,
all fairly small countries, mainly in Northern Europe.
!
Lower risk countries – France, Germany, Italy – large countries (but with little data on risk).
!
Other countries cannot be classified, for lack of evidence.
In conclusion, we suggest that this country classification is best regarded as a hypothesis – the basis for
further research to test the classification and amend as appropriate. Usefully, however, it suggests that, for
some countries, being new to the internet means gaining new risks while for others, having gained high
internet penetration carries with it high risks. The anomalies are also interesting – Sweden has high internet
penetration but only average risk, suggesting perhaps an effective level of safety awareness.
Comparing now across risks instead of across countries, the same findings reveal that some risks are more
prevalent than others, though variation across European countries is considerable. Setting aside sending
bullying/harassing messages (where there are data for too few countries), the order of risks is roughly:
1.
Giving out personal information: the most common risk – estimates around half of online teens, with
considerable cross-national variation (13% to 91%).
2.
Seeing pornography: second most common risk at around 4 in 10 across Europe, but there is
3.
Seeing violent or hateful content: third most common risk at approx one third of teens and, apart from a
figure of 90% in Ireland (and 51% in Poland), a fair degree of consistency across countries.
4.
Being bullied/harassed/stalked – generally around 1 in 5 or 6 teens online, though there is also a group
of high risk countries here (Poland, perhaps Estonia) and one low risk country – Belgium.
5.
Receiving unwanted sexual comments - only around 1 in 10 teens in one group of countries (Germany,
Ireland, Portugal) but closer to 1 in 3 or 4 teens in Iceland, Norway, UK and Sweden, rising 1 in 2 in
Poland.
6.
Meeting a online contact offline – the least common but arguably most dangerous risk, there is
considerable consistency in the figures across Europe at around 9% (1 in 11) online teens going to such
meetings, rising to 1 in 5 in Poland, Sweden and the Czech Republic.
7.
The miscellaneous category of findings also listed above indicates the range of possible risks still to be
researched in comparative perspective – self-harm, race hate, commercial exploitation, and many more.
From this survey of findings, several conclusions may be tentatively drawn.
!
First, there are considerable cross-national variations in the incidence of risk, although it is hard to
discern systematic cross-national patterns across all risk types.
!
There seems to be more cross-national variation in the more common risks, and more homogeneity for
the less common risks.
!
In several countries, some measure of distress or feeling uncomfortable or threatened was reported by
15%-20% of online teens, this suggesting, perhaps, the numbers for whom risk poses a degree of harm.
31
!
Poland is a striking outlier – reporting high levels of risk across several categories, and being highest for
seeing porn, being bullied, receiving unwanted sexual comments, second highest for stranger danger,
and third highest for giving out personal information.
!
Further, in some countries it is particular risks that are stand out, but they are not high risks across all
risks e.g. Ireland for seeing violent and hateful content and giving out personal information, Czech
Republic for giving out personal information, Estonia for being bullied.
!
The only country that is somewhat (comparatively) a low risk outlier on a few items is Italy – on porn,
seeing hateful content, although the actual figures are not so striking as the high risk outliers, and this is
partly because the samples surveyed were younger (7-11 year olds).
The research gaps are considerable, for there is little or no research evidence across Europe regarding
some forms of risk to children online. Specifically, there is little on commercial risks (either those which direct
commercial/advertising content to children or those which track their online activities or collect personal
data). There is also little research on risks associated with exposure to certain values. Although there are
scattered studies of the incidence of exposure to racist content, though these are too few to compare crossnationally, and thus they have been combined with hateful content. We have found few or no studies on selfharm (e.g. cutting, suicide, pro-anorexia) or on inappropriate forms of persuasion or misleading advice. Only
recently have there been some studies of children not as victims but as perpetrators, focusing on bullying
and sending unwanted sexual messages.
Undoubtedly, there are some difficulties in researching these topics – both practical and ethical, but
nonetheless, the attempt should be made. The measures asking not about specific risks but about children’s
possible distress or fear associated with these experiences provide a helpful indication of possible harm, and
require further investigation to understand the duration and severity of such responses.
Some of the high reports of risk – in Estonia, Poland, Czech Republic – require urgent awareness-raising.
Similarly, the advent of new forms of online activity – e.g. social networking – points to the need for urgent
new advice to children and young people. As estimates for now-familiar risks continue to be substantial,
these too require continued attention to keep them in children’s minds.
2.2.3
Positional factors influencing children’s online opportunities and risks
We now turn to the hypotheses that link the demographic factors of age, gender and socioeconomic status to
findings for children’s experience of online risks.
Hypothesis H2.2.1: As children get older they are exposed to a greater amount and range of online risks.
It is commonly supposed that as children get older they are exposed to a greater amount and range of online
risks (although, as is also supposed, this may matter less than for younger children as older children are also
more mature and capable).
Of the 21 participating countries, only eight had evidence to support this hypothesis (Belgium, Estonia,
Iceland, Norway, Poland, Portugal, Spain and the UK) and three had contradictory evidence (France,
Germany and the Netherlands). The remaining countries (Austria, Bulgaria, Cyprus, Czech Republic,
Denmark, Greece, Ireland, Italy, Slovenia and Sweden) had no data to present.
!
As the Austrian report noted, this is because they, like many countries, lack direct evidence of risk for
children younger than teenagers – in other words, lack of evidence does not mean the hypothesis can
be rejected. On the other hand, as they also point out, across the age range from 11 to 18, they – again
like many countries - have evidence of risks encountered across the teenage years.
!
Also in support of the hypothesis, research on Estonia shows that “Cyber-bullying, stalking, harassment”
has been experienced by 42% of 13 to 14 year old children, 34% of 11 to 12 year olds, 37% of 9 to 10
year olds and only 11% of 6 to 8 year olds. Also, older children communicate more with strangers on the
Internet: 49% of 13 to 14 year olds have communicated with strangers (8% with adult strangers) while
only 7% of 6 to 8 year olds have done so (2% with adult strangers; (Turu-uuringute AS, 2006).
!
Similarly in Iceland, as children grow older they are more likely to have met a stranger on the internet
who asked for personal information about themselves (12% of 9 year olds compared to 30% of the 15
32
year olds). They are also more likely to have received unwanted sexually explicit messages (16% of 11
year olds compared to 26% of 15 year olds). Moreover they are more likely to have stumbled into
websites with pictures of naked people or porn sites (44% of 11 year olds compared to 63% of 15 year
olds). And in Spain, a range of content risks (promoting violence, war, terrorism, pornography, etc) are
more commonly experienced by 15-17 year olds than by 12-14 year olds, as is the likelihood of meeting
in person someone first met online - 15.1% of adolescents between 15 and 17 years of age do this, as
do 8.2% of younger children, ages 12 to 14 (Tezanos, 2006).
On the other hand, some research finds contradictory results.
! In France, research shows that as children get older they are less likely to participate in chat rooms
(comparing 15-17 year olds with those aged 18+), resulting in a decreased chance of facing some risks.
! In Germany, there was no evidence directly addressing the relation between age and risk exposure, but
findings on problematic films on mobile phones suggest that children aged 12-17 own more of these films
on their mobile phones compared with those aged 18 and above.
! Findings from the Netherlands suggest that children aged 12-14 encounter greater risks when using
online technologies because they’re experimenting with their identities and are more prone to receive
negative feedback of their profiles, thus endangering their well being.
On balance, however, it is concluded that older teenagers do encounter more online risks. To qualify this, it
may be that for some risks, the oldest teens have learned how to avoid such risks, while younger teens are
the most likely to be sensation-seeking or deliberately risk-taking. Also, the phenomenon of very young
children using the internet is too recent for a strong evidence base, raising new and pressing questions.
Part of the explanation for the generally positive association found between age and risk is offered by the UK
Children Go Online study (Livingstone & Bober, 2004). This conducted a path analysis showing that the
positive correlation between age and risk among 12-17 yr old internet users (r=0.26, p<0.01) may be
because older children engage in more online opportunities and this, indirectly, leads them to experience
more risks. In short, older teens do more online of a beneficial nature, and this indirectly leads them into
more risky experiences.
H2.2.2: As younger children gain online access they are increasingly exposed to online risk.
It is possible that as ever younger children gain access to online technologies, they may be particularly
vulnerable to risk – gaining access before developing the requisite skills or maturity to cope with what they
encounter. The above analysis also suggests that as younger children gain online access, they will
encounter more online risk precisely because they will take up more opportunities than at present.
But direct research is yet to be conducted to explore the specific risks faced by younger children. And
unfortunately, even where survey findings across the age range are available, they are rarely analysed so as
to address this issue. Across Europe, most countries had no findings examining the interrelations among
children's age, online access and risk exposure. Some had findings to suggest that, as younger children are
more closely monitored by their parents, risks are less problematic when encountered among young children
(e.g. Iceland), but this point relates more to the consequences of this hypothesis, if supported, than to the
merits of the hypothesis itself.
H2.2.3: There are gender differences in the range/types of uses/opportunities
Data from thirteen countries support the hypothesis that there are gender differences in the range and type
of children’s online activities: Bulgaria, Estonia, France, Germany, Iceland, Ireland, Italy, the Netherlands,
Norway, Poland, Portugal, Spain, Sweden.
The hypothesis is challenged in only two countries which provided mixed evidence – i.e. research results
which challenge as well as support the hypothesis (Greece, UK). Limited evidence was available in Czech
Republic. No evidence was available from four countries: Austria, Denmark; Slovenia, Belgium.
The 2005/6 Norwegian SAFT survey of 888 9-16 year olds shows a fairly typical pattern of online activities in
terms of gender (see table 2.8).
33
Table 2.8: Online activities of Norwegian children between 9 and 16 years
What kind of things do you do on the internet?
Play games on the internet
Do homework
Download music
Chat in chat rooms
Send and receive email
Search for information (other than schoolwork)
Surf for fun
Instant messaging
Visit news sites
Make personal website/blogging
Visit fan sites
Publish pictures or information
Download software
Visit sites for hobbies
Shop or make a purchase
Watch pornography
Boys (%)
84
52
57
47
44
38
41
32
32
22
21
15
23
13
16
15
Girls (%)
59
64
54
48
50
40
29
30
27
24
23
20
8
14
8
3
Source: SAFT Children’s Survey (2006). Findings presented at the Stakeholder Event for the German launch of EU Kids
Online, University of Hamburg, December 2006. See also http://www.saftonline.no and http://www.saftonline.org
Overall, findings from across Europe suggest that while boys and girls enjoy many similar activities, there are
some common gender differences as follows:
!
In general, boys are involved in a wider range of online activities and have different preferences than
girls, particularly in types of downloads and gaming activities.
!
Boys prefer sport-oriented and action games. Girls favour adventure, party and mind-challenging games
and self-expression.
!
There is also a difference in internet surfing in most countries. Girls are more likely to search for
information for educational purposes, boys for entertainment purposes.
!
Differences are apparent also in the use of the internet for communication, with more girls than boys
being regular users of email, MSN and blogs.
!
Girls are more likely than boys to publish photos of themselves. They access a wider range of usergenerated content than boys.
There is little cross national variation in findings. Following the above finding that gender gaps in internet use
disappear in countries wit a high level of use (see table 2.5), it may be hypothesised that gender differences
are greatest in the early days of internet diffusion in a country, where social expectations and access
provision are highly gendered, and that with familiarity and embedding in daily life, these gender differences
diminish. However, too little evidence is available to examine this hypothesis for the use of specific services.
Only in Poland is email more widely used by boys than girls. All other countries report girls’ higher use of the
internet for communication purposes than boys. Things are changing fast, also: for example, although
computer games have traditionally been targeted at boys, recent years have seen a greater number of
games appealing to girls and this is increasing girls’ game-playing.
Thus it is concluded that while boys and girls enjoy a range of online opportunities, there is clear evidence of
gender differences in online activities and preferences. Girls prefer activities that involve communication,
content creation and collaboration. Boys prefer competition, consumption and action. As yet, too little is
known regarding the relatively new phenomena of social networking, online and multi-user gaming and other
web 2.0 activities.
The relationship between gender, age and internet activity needs to be investigated more fully.
34
!
For example, in Germany, it appears that gender differences increase with age: while there are minimal
gender differences in children’s (6 to 13 years) internet surfing (KIM, 2006), there are very significant
differences between teenage boys and girls’ internet activities; for most internet activities the
percentages of the boys are higher than that of girls; and only e-mail and information seeking for the
school and job are used more frequently by girls than by boys (JIM, 2007).
!
A similar gender gap appears in findings for UK teenagers compared with younger children: among
younger children, there is little if any gender difference in opportunities taken, but by the early to midteens, by which time the number of opportunities taken up is expanding, a gender difference has
opened up, with the girls reaching a plateau at around 6 or 7 opportunities (from a list of 31) while boys
continue to expand their online opportunities until they too reach a plateau by the age of 16-17 years.
We need to build on this research base. There is limited data in some countries and rich data in others,
particularly those that have used a mix of qualitative and qualitative methods. We also need to examine
children’s motivations for using the internet for various activities and the influence of friends and peers on
internet activities/opportunities. We do not know, further, whether how people learn to use the internet
influences (either constrains or expands) their online activities.
H2.2.4: There are gender differences in the range/types of risks
Even more than opportunities from online activities, it is popularly supposed that there are gender
differences in the extent and nature of children’s exposure to online risk (indeed, to risks of all kinds).
As the findings showed, there are indeed gender differences in exposure to risks. Fourteen countries
provided research results supporting the hypothesis. There is limited evidence available in Belgium but what
is available supports the hypothesis. There was no evidence available in six countries: Austria; Bulgaria;
Czech Republic; Denmark; Slovenia; Sweden.
Overall, it may be concluded that:
!
!
Boys are more likely
o
to seek out offensive or violent content,
o
to access pornographic content or be sent links to pornographic websites,
o
to meet somebody offline that they have met online,
o
to give out personal information.
Girls are more likely
o
to be upset by offensive, violent and pornographic material,
o
to chat online with strangers,
o
to receive unwanted sexual comments,
o
to be asked for personal information but to be wary of providing it to strangers,
! Both boys and girls are at risk of online harassment and bullying.
National findings for gender differences in types of risk are summarized in table 2.9.
35
Table 2.9: Summaries of national reports on types of risks
Form of risk/
role of child
online
Aggressive
content (child as
recipient):
…. Seen violent
or hateful
content
Aggressive
contact (child as
recipient):
…. Been
bullied/
harassed/
stalked
Aggressive
conduct (child as
actor):
… Sent
bullying/
harassing
messages
Sexual content
(child as
recipient):
… Seen
pornographic
content
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Sexual contact
(child as
participant):
Received
unwanted
sexual
comments
Sexual contact
(child as actor):
Chat online
with strangers
Met online
contact
(stranger)
offline
•
•
•
•
•
•
•
•
•
Given out
•
•
France: Boys more likely to download harmful content
Germany: More boys than girls have seen a fight filmed on mobile phone. More boys
have violent videos on their mobile
Ireland: Boys are three times more likely than girls to have visited hate sites.
Norway: Mainly boys visit websites with offensive content.
Poland: From age 10 a higher percentage of boys than girls is at risk of exposure to
illegal and offensive material
UK: Boys more likely to seek out violent or gruesome content.
Belgium: Girls have felt more threatened than boys
Germany: More girls than boys (12-19 years) have met unpleasant people in a
chatroom
UK: Girls are more likely to have been bullied online.
Estonia: Cyber-bullying, stalking, harassment is a bigger risk for boys than girls
Ireland: More boys than girls (9-16 years) have experienced online harassment
Italy: The risk of harassment is higher for boys than for girls
Italy: More boys (7-11 years) than girls have sent harmful content through the internet
or mobile phone
Cyprus: All girls have seen nudity on the internet
Iceland: Girls are much more likely than boys to say that they experienced discomfort
when looking at pornography.
Poland: Girls are more likely to be exposed to erotic material in emails and
chatrooms. Girls find sexual and erotic content more shocking than boys.
The Netherlands: More girls than boys are upset by sexual images on the Internet
Germany: More boys than girls have viewed pornographic material online
Poland: Boys are more likely than girls to insert key words in search engines related
to sex and erotica.
Iceland: Boys are much more likely than girls to have visited pornographic websites.
Spain: Boys are more likely than girls to receive Internet links to pornographic web
pages
UK: Boys are more likely to encounter online pornography, both accidentally and on
purpose
France: Many girls visiting chat rooms say they had contacts with older boys or men,
explicitly talking about sexuality
Iceland: Girls more likely to have received unwanted sexually explicit messages than
boys. They are also more likely than boys to have been asked to send naked picture
of themselves
Norway: Mostly boys who receive unwanted sexual comments.
France: Young teenage girls may hide their identity in chat rooms and pretend they
are boys or older girls, in order to learn more about sexuality or online flirting.
Portugal: More girls than boys chat online with strangers
France: More girls than boys chat online to strangers
Norway: Boys are more likely to have face to face meetings with people they have
met on the net.
Iceland: Boys are more likely to have face to face meetings with people they have
met on the net.
Portugal: A higher percentage of boys than girls have contacted people off-line they
have met on-line
Spain: More boys than girls meet people offline
Estonia: strangers in online forums are more interested in getting to know the real
36
personal
information
•
•
•
•
Miscellaneous
other risks
•
•
names of girls.
Germany: More girls than boys (12-19 years) have been asked in a chatroom by
strangers for their address, telephone number and name. Fewer girls than boys
provided the information.
Greece: Fewer girls than boys would share personal details with a stranger
Ireland: Girls (9-16 years) more conscious than boys of being contacted by strangers
when submitting personal information.
Italy: More boys than of girls reported giving out false information about another
person
UK: Boys more likely to give out personal information online
Poland: More boys than girls receive links to websites, e-mail, mail addresses
telephone numbers providing informing on how to access illegal or inappropriate
contents
Although this table suggests a considerable diversity of gender differences in exposure to risk, it does not
provide the basis for a clear classification of countries on the basis of gender differences or otherwise.
More comprehensive investigation of exposure to and perceptions of risk is needed, particularly boys’
tendency to seek out violent and pornographic material. Links between boys’ games content and playing
(which often have violent or sexual content) and exposure to offensive and sexual online content would be
worth exploring. It would be important also to examine awareness of risks and coping strategies among girls
and boys.
Hypothesis H2.2.5: There are inequalities in use/opportunities as a consequence of inequalities in SES
The well-established debate over the digital divide justifies the hypotheses that there are inequalities in
children’s use of and opportunities gained through the internet as a consequence of differences in
socioeconomic status (SES).
Though evidence was lacking in several countries (Austria, Cyprus, Czech Republic, Denmark, Greece,
Iceland, Ireland, Italy, Poland, Portugal, Slovenia), there is general agreement throughout many European
countries that SES and the type of use or opportunities are correlated. Different variables are examined in
several countries, here categorised in terms of measures of frequency of use or type of use.
Frequency of internet use:
!
In some countries (Estonia, France, Sweden), there is little difference for children of different
socioeconomic status.
!
The different ways of measuring use complicates these simple conclusions, however. For example, in
Estonia, among 15 to 19 year olds, a positive correlation exists between household income (per family
member) and variety of computer use (the number of different activities youngsters engage in). On the
other hand, there is no significant correlations between household income and the frequency of
computer/Internet use, or between household income and the amount of time spent on computer use at
home (MMM Project, 2005).
!
By contrast, a positive association is found in Norway and the UK. Children whose parents have more
education or who are of a higher class use the internet more.
!
Belgium and the UK also report that upper class children started to use the computer younger.
!
Qualitative research in France (Fluckiger, 2007) shows that inequalities in use could result from the
transmission of cultural capital, as parents with a higher education encourage their children to use the
internet more and more widely.
!
Surveys in both France (Pasquier, 2005) and the UK (Livingstone & Helsper, 2007) show that although
there are inequalities in possession of a computer or access to Internet as a consequence of
inequalities in SES, among teenagers who have access to the Internet, there is little significant
difference in the frequency of use.
Type of use:
37
!
Several countries report that upper class children use internet more often for school (Netherlands), to
get information about important questions or to do their homework (Spain), or to contribute to message
boards, vote or sign a petition online or visit civic sites (UK); children from “theoretical” schools are more
likely to make use of scientific and political information on the Internet (Sweden).
!
On the other hand, downloading music is more frequent among children from lower class households
(France, Netherlands, UK).
!
Even in a country considered relatively egalitarian, like Norway, the differences are notable. Children in
families where parents have higher education use the computer more than others. The difference is
largest concerning homework, e-mail and gathering information, but also significant when it comes to
image processing and music. Concerning chat, programming and computer games, the difference is
small or insignificant. Children with parents of lower education use TV-console games more than do
others. There is a positive correlation between how much time the children spend on the computer and
how good grades they have, and their parents’ educational level. (30)
!
Overall, all countries agree that lower class children use Internet for leisure information, downloading
content, and fun. Upper class children tend also to have uses related to school, civic or political
information.
!
For example, in Spain, 75% of the young people interviewed from the higher social status group said
they generally browsed the Internet to obtain information when doing school work or homework; only
43% of the lower social status group do this (Tezanos, 2006).
!
Apart from UK and France (where working class children are more likely to use chat rooms), the reports
do not suggest any association between SES and the use of communication tools.
Does the frequency of use and type of use depend on SES? The answer is positive for some national
reports, negative for some others. A possible explanation is that frequency of use is sometimes asked
whatever the place – thus use at school is included, which may lower or even cancel the effect of SES.
Hypothesis H2.2.6: Since most children make the broadest and more flexible use of the Internet at home,
they will also encounter more risk from home than school use.
This hypothesis is particularly important in terms of directing safety awareness initiatives – whether to
parents or to teachers – and in framing advice to children directly (whether phrased in terms of school uses
or home uses, positioning either or both of parents and teachers as advisors). However, it is not
straightforward to investigate, since it combines several statements that have to be investigated:
1.
2.
3.
4.
How many children use the Internet at home and at school?
How often and for how long do they use the Internet at home and at school?
How broad is the range of Internet services, which are used at home and at school?
How likely is it to encounter risks when using the Internet at home and at school?
The Eurobarometer (2005/06) survey (EC, 2006) allows the investigation of the questions 1 and 4, although
only by relying on parents’ answers (problematic since parents cannot know about their children’s Internet
use at other places, including school).
For question 1, Table 2.2 (see above, chapter 2.1) has shown how many children use the Internet at home
and at school. On the EU level this first indicator for potential risks shows no difference for Internet use at
home (34%) and at school (33%). However there are countries with substantially higher figures for use at
home (Belgium, Denmark, Germany, Italy, Slovenia, Spain, Sweden, and The Netherlands) on the one hand,
and countries with higher figures for use at school (Bulgaria, Czech Republic, Greece, Poland, Portugal, and
the United Kingdom) on the other hand. The latter group includes Central and Eastern European countries
as well as Southern countries with a generally low Internet use; the only exception here is the United
Kingdom.
The second step of the analysis asks whether the children’s parents report that their child has encountered
any harmful or illegal content when using the Internet at different places. Table 2.10 shows that the parents
38
observe more risks at home than at school; this is true for the European Union as a whole as well as for all
countries involved in the EU Kids Online project. These findings clearly support Hypothesis 2.2.6 (although
as noted above, parents’ knowledge of their children’s internet use at school is limited).
A more specific approach to this question is based only on those children who – according to their parents –
use the Internet at home AND at school. On this basis, the hypothesis is clearly supported: on the European
Union level, the figure for risks at home is much higher (23%) than for risks at school (7%).
Table 2.10: Encounters with harmful or illegal content when using the Internet at home and at school
(in per cent of children who have used the Internet)
Austria
Belgium
Bulgaria
Cyprus
Czech Republic
Denmark
Estonia
France
Germany
Greece
Iceland
Ireland
Italy
Netherlands
Norway
Poland
Portugal
Slovenia
Spain
Sweden
The UK
EU 25
At home
7
20
3
5
11
22
17
11
7
10
nd
6
8
31
nd
11
8
20
14
34
12
12
At school
5
3
4
3
7
14
4
0
3
3
Nd
1
8
9
Nd
6
2
9
2
16
4
5
Source: Eurobarometer 64.4 – Special No. 250: Safer Internet, December 2005;
As a final step of the analysis, all places where children use the Internet can be qualified regarding the
likelihood to encounter harmful or illegal content. The following analyses are based only on those children
who actually use the Internet at the respective place. Table 2.11 shows the figures on the European level.
Because of the fact that most children use the Internet at home, this place clearly seems to be the least
secure: 12 per cent of the parents say that their children have encountered harmful or illegal content when
using the Internet at home; this is by far more than for the other places.
However, if one takes into account the number of children who actually use the Internet at the different
places, the picture changes: Still the highest figure can be observed for Internet use at home; 18 per cent of
the children who use the Internet at home have encountered illegal or harmful content at this place. A
similarly insecure place (17%) is the Internet café. Internet use at a friend’s house is in third place (12%),
whereas somebody else’s house (8%) as well as schools (7%) and particularly libraries and other public
places (3%) are regarded as more secure.
These findings cannot be calculated on the country level, because for most of the countries the number of
cases is too small. Looking at the figures does not provide any indication for substantial differences between
the countries in this respect.
39
Table 2.11: Likelihood to encounter harmful or illegal content at different places (EU 25; in per cent of
children, a) who use the Internet at any place or b) who use the Internet at the respective place)
At home
a) all children who
use the Internet
b) children who use
the Internet at the
respective place
12
N=3791
18
At
school
5
n=3791
7
N=2590
n=2514
At a
friend’s
home
4
n=3791
12
At someone else’s
home
1
N=3791
8
At
Internet
cafés
1
n=3791
17
At libraries or
other public
places
0
n=3791
3
n=1218
N=348
n=199
n=285
less than 18 years.
Taken together, the Eurobarometer data – as far as they include the relevant variables – clearly support
Hypothesis 2.2.6. From the parents’ perspective the likelihood of encountering harmful or illegal content is
substantially higher when children use the Internet at home than at school. In addition, Internet use at a
friend’s home and at Internet cafés is regarded as more dangerous than at school. This provides a
reasonable basis for targeting safety awareness information to parents and to children accessing the internet
at home.
Beyond this overall picture, there is little available research that maps children’s reported uses and risks
online across multiple locations of use, on a country by country basis. Further work is therefore needed that
surveys children directly, using sufficient sample sizes and asking about uses and risks across the various
locations where children access the internet. A few national findings are worth noting, however, to gain a
picture of the degree of variation likely across Europe.
For example, in Bulgaria, the Eurobarometer survey finds that more children use the Internet at school (12%)
than at home (8%), this explaining why the likelihood of encountering risks is 4% for use at schools and 3%
for use at home. As the national report comments, risks at home arise because of parental ignorance of or
inexperience with the internet, and their belief that what children are doing at home is what they have
learned at school. The results of the national study performed in 2006 show that parents think their kids are
safer in cyber clubs than in the street and therefore they encourage children to attend. Parents usually are
not aware of what their kids are doing in the Internet and rarely could prevent the risks (with few exceptions).
Qualitative research in France (Fluckiger, 2007) shows that at school, computers and the Internet are
underused. Moreover, uses at school are very restrictive, so that pupils can not surf the Web freely, chat or
visit blogs. Therefore, risks encountered at school appear very limited. Similarly in Ireland, children use the
internet more frequently from home and are less likely to be supervised, whereas the Schools Broadband
Programme has filters, social networking sites are banned at school, schools employ secure email systems,
and less time is available for internet use as activities tend to be structured around the curriculum.
While further national data are largely lacking, those that are available further support the conclusion that, in
general, online risk is greater at home than in school.
Overall, it seems that, in general, both frequency of use and type of use are influenced by SES – parental
resources (economic, cultural, educational, social) resulting in some children benefiting more from the
internet than others. But there are several exceptions, and gaps in the available evidence. Too few studies
discriminate use at school (relatively more equal) and home (more unequal).
Research question R2.2.3: Are there SES differences in children’s exposure to risk?
The relation between the digital divide (i.e. the risk associated with not accessing the internet, of being
excluded) and online risk exposure (the risk associated with accessing the internet) has been little examined
but is nonetheless important, hence this research question.
If it is supposed, as seems plausible, that parents are differentially resourced to manage online risk
exposure, and that children are already – in their offline lives – differently at risk (or ‘vulnerable’), this lack of
attention to questions of SES is a concern in the research field. It is strongly recommended that all future
research examines the differential consequences of internet access and use and, indeed, safety awareness
and risk responses, for households of different socioeconomic status.
40
In most countries where research was collated, there was little information available regarding
socioeconomic status (SES) – including Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark,
Greece, Italy, Norway, Poland, Portugal, Slovenia, Sweden, The Netherlands. Insofar as findings are
available, the evidence in each country points to a correlation between SES and exposure to risks, with the
exception of Iceland. Most of these findings concern content and contact risks. In the main, it seems that
lower class children are more exposed to risk online.
! For example, in Spain, Germany and France, there is evidence that working class children are more likely
to encounter, receive or view pornographic or violent material, either on their email, Web browser or
mobile phone.
! Spanish research shows that only 2.7% of young people from the highest social group state that they
have received violent contents through the Internet, as opposed to the 60.3% of young people with a
middle social status; similarly, 3.9% of young people from the very highest social group accidentally
access pornographic pages, while 26.2% of those with a high social status and 61.2% of those with a
middle social status do so. Similar data correspond to young people who receive pornographic pages
through the Internet or messages: 5.5% belong to the highest social status, 26.6% are from the high
social status group and 5.8% have a middle status. Pornographic pages received from someone known:
8.2% from the highest status, 28.6% from the high social status and 57.1% from the middle social status.
Surprisingly, when asked whether or not they know someone who visits pornographic pages habitually,
the highest social status gave an affirmative response in 61.5% of cases, those with a high social status
24%, the middle social status group 28.7% and those of the lowest social status, in 14.3% of cases
(Tezanos, 2006).
! In Germany, a study of mobile phones and risks indicates that less educated teenagers own violent video
films on the mobile phone to a larger extent than better educated teenagers:
General-education secondary school
11.9%
Secondary modern school
6.4%
Grammar school
2.1%
Source: (Grimm & Rhein, 2007). Base: teenagers with mobile phones (12-19 years, n=752).
!
In France also, the findings suggest that working class children are more likely to talk to strangers on the
Internet, or meet strangers they have met online. Thus Pasquier’s qualitative and quantitative research
between 2001 and 2005 (Pasquier, 2005) found in France that 49% of working class high school pupils
talk to people they never met before, compared to 41% of middle class and 26% of upper class
teenagers. She argues that working class teenagers, more than others, say they are interested into
chatting because chatting is seen as an occasion to demonstrate one’s ability to master the specific
language in chat rooms, and that is particularly valued among under-privileged teenagers.
It can be concluded that those who belong to higher SES groups are generally exposed to fewest risks.
Further, middle SES groups experience more risk and lower SES groups experience the most. It seems
likely that several factors are at work here, with the relatively lesser access of the lowest status groups
resulting in less exposure to risk, thus complicating the correlation between SES and risk. It is noteworthy
too that the Irish report finds that only 41% of lower SES parents monitor their children’s internet use,
compared with 81% of other groups, and that children from lower SES groups are more likely to have access
to computers and the internet in their bedroom than higher SES groups (Downey, Hayes, & O’Neill, 2007).
Lower parental monitoring may, it seems, be associated with – possibly result in – greater exposure to risk
among children.
Some results are surprising. The UK reports only a small difference, and Iceland claims there is no SES
effect. Internet is widely spread in Iceland, and, according to the report, very little socially differentiated,
which could explain this result. However, UK is a highly stratified country, and the existing but small
difference in the exposure to risks requires a different explanation. It should be noted that the relatively small
difference in use in the UK exists only after controlling for access. In other words, access to the internet is
highly stratified in the UK but, once access is achieved, SES makes little measurable difference to the
frequency or nature of risks.
41
For future research, it will be important to systematically differentiate the various dimensions of SES:
household income, level of education and profession of parents, so that the specific resources of households
- social capital, economic capital, cultural capital – can be distinguished as factors exacerbating or
ameliorating risk. SES differences in parenting style are yet to be well understood. Last, other forms of social
inequality – e.g. the case of migrants and ethnic minorities – should be included in research designs.
In terms of safety awareness, these findings suggest the value of targeting interventions at working class
children especially.
2.2.4
Consequences of online risks and coping
Research question R2.2.4: Is there evidence showing the consequences of online risks or evidence showing
how children cope with online risks?
The Safer Internet plus programme’s public consultation (Safer Internet plus programme, 2007), summarised
and discussed at the 2007 Safer Internet Forum, highlighted several key conclusions, of which the first – that
risk and safety should be addressed in the context of the ‘overwhelmingly positive potential of the internet’
has been addressed in the section on online opportunities. The second was that since ‘a risk free internet for
children and young people is an illusion’, the focus should be on risk avoidance, coping with risk and media
literacy.
We have begun to address this above, when examining whether increased online skills reduced risk
exposure. But beyond the question of exposure is the important matter of coping. Once exposed to risk, how
do children respond? In psychological research, this question is being framed in terms of adolescents’
development of ‘resilience’.
Thus far, however, little is known of children’s abilities to cope with, or their resilience towards, online risk.
Some findings do exist, however, and these are often promising, for they tend to suggest that such risks as
children do encounter may be brushed off, or disregarded, by the majority of young people.
This leaves two crucial questions.
!
First, methodologically speaking, can children be asked to self-report on harm with reliable results (in
other words, might they be harmed in ways they cannot or choose not to describe when asked by a
researcher)?
!
Second, are some children particularly vulnerable to online risk, even though the majority appear to be
relatively unaffected?
These two questions must remain for future research.
In what follows, we note the available evidence for children’s responses to online risk. The Eurobarometer
survey 2005/06 (EC, 2006) includes two questions directly related to risks, and combining these provides
some indication of coping. Firstly, parents were asked whether their child has ever encountered harmful
st
content on the internet (see table 2.12 1 column). Secondly, they were asked whether they think their
nd
children know what to do if a situation in the internet makes them feel uncomfortable (see table 2.12, 2
column). This item can be interpreted as an indicator for parents’ trust in their children’s ability to cope with
online risks.
On the European level, 31 per cent of the parents say that their child has encountered harmful content on
the internet, and 66% of parents say their child knows what to do in such situations.
42
Table 2.12: Children’s encounters with harmful content and their ability to cope with internet related
risks
(parents’ answers; per cent of children who use the internet)
1) Child has
encountered
harmful content
(%)
EU 25
Bulgaria
Estonia
Sweden
Slovenia
Poland
Czech Republic
Netherlands
Austria
Denmark
Spain
Portugal
Greece
Belgium
Ireland
Germany
Italy
United Kingdom
Cyprus
France
30,8
58,2
57,5
55,5
55,4
48,8
48,4
43,2
40,8
39,6
36,7
33,7
30,5
26,3
25,5
23,3
22,1
21,9
20,7
19,2
2) Child knows what to
do in situations, which
make them feel
uncomfortable
(%)
66,0
48,4
44,1
65,9
61,1
55,6
58,0
72,8
70,1
67,6
50,0
46,7
50,8
67,0
64,8
71,0
69,2
75,3
72,4
69,5
Correlation between
encounters with harmful
content and coping
(within countries)
(r)*
-,02
,12
-,15
-,25
-,02
,02
,02
-,25
,19
-,15
,13
,16
,07
-,20
,13
,03
-,10
-,09
,32
,00
less than 18 years.
*) Shaded cells indicate significant (p<.05) correlations.
Figure 2.5 shows the aggregate results for the EU member states. Evidently, there is a negative correlation
13
between the two indicators across countries : the higher the percentage of parents who claim their children
have encountered harmful content, the lower the estimated ability of children to cope with these potentially
harmful encounters.
13
The Pearson correlation across the 19 countries is r=-.52, p<.05.
43
Figure 2.5: Countries by encounters with harmful content and children’s ability to cope with
situations, which make them feel uncomfortable
(based on parents’/guardians’ answers, in per cent of children who use the internet)
% Child knows what to do in situations, which make them
feel uncomfortable
80
FR
70
CY GB
FI
DE
IT
BE
LU
AT
DK
SE
SK
IE
60
NL
CZ
MT
50
GR
RO
PL
ES
PT
LV
SI
BG
HU
EE
40
2
R = 0,25
LT
30
20
10
0
0
10
20
30
40
50
60
70
80
% Child has encountered harmful content
less than 18 years.
Some caution is needed in interpreting this correlation. It may be that in low risk countries, children have
indeed learned to cope; but it may also be that in low risk countries, parents are unaware of the need to cope
and so overestimate their children’s abilities. Similarly in high risk countries, children may really be less able
to cope or, possibly, in high risk countries parents are more aware of their children’s need to cope.
These figures suggest that the highest risk/lowest coping countries are Estonia and Bulgaria, followed by
Poland and the Czech Republic – clearly a priority focus for future safety awareness initiatives. On the other
side of the spectrum, a group of seven countries (Belgium, Cyprus, France, Germany, Ireland, Italy, and the
UK) combine low risk and a high ability to cope.
Note that this classification of countries in terms of risk is somewhat at variance with that offered earlier,
based on far more national research with children. The above findings are, here, interpreted less in terms of
actual risk and more in terms of perceived risk by parents, especially as this reveals a gap between parental
perception of risk and parental assessment of child’s coping abilities.
Further evidence of the pattern of parents’ perceptions of online risks and their children’s ability to cope with
them can be derived from the within countries correlations between these two indicators as shown in the
rd
14
above table 2.12 (3 column). On the European level the correlation across individuals is almost zero;
however on the country level we find several countries with a significantly positive correlation (Austria and
Cyprus) as well as other countries with significantly negative correlations (Belgium, Denmark, Estonia,
Netherlands, Sweden). This means, in the first group parents, who think their child has encountered harmful
content, are more likely to believe in their child’s ability to cope with this than parents, who do not think their
child has encountered harmful content. To the opposite, parents in the second group, who think their child
has encountered harmful content, are less trustful regarding their children’s ability to cope with risks. This
pattern is not easy to interpret; it emphasizes the relevance of thorough analyses within countries, because
relevant relations between core variables seem to be substantially different.
How do children cope with online risk? In Table 2.13 below, we cull findings on children’s responses to risky
experiences from the national reports that form the basis for this comparative analysis.
14
Here we find an impressive example that correlations on the between-countries-level must not be mixed with
correlations on the individual level. Whereas on the between-countries-level there is a clear negative correlation (see
above), this cannot be found when calculating correlations on the level of all individual respondents.
44
Table 2.13: Summaries of national reports on children’s abilities to cope with risks
Belgium
•
Pupils fairly confident they can manage the risks of online contact with strangers.
Denmark
•
When visit hateful websites, 24% don’t think much about it; 23% get upset; 7%
thinks it’s funny; 6% agree with the content; 5% thinks it’s cool
39% of children ignore it when they come across violent content. 19% of children
have visited harassment sites. Most of them don’t give it much thought or they might
think that it’s fun or cool to agree, with the content on the site. A third of the children
passed the link on to friends.
Qualitative research reveals children’s strategies for caution online e.g. revealing
personal info gradually when trust; meeting stranger in open places/with friends;
motivations for risk-taking also explored – learning to flirt, testing identity and
relationships, exploring adult world
Qualitative Eurobarometer – children (9-10 and 12-14) generally conscious of online
risks and take measures to reduce them, following parental advice; children inform
friends more than parents and discuss how to cope.
Of those children who had seen a website with pornographic content 41% told a
friend 17% told an adult and 53% told no one about it.
Of those children who had seen a website with pornographic content 42% never
visited that site again and another 43% ignored the site.
Of those who had seen websites with pornography or violent content 47% said they
did not think about it, 33% said they found it unpleasant, 18% said they wished they
had not seen it, 13% said they found it odd and 10% said they thought it was cool.
Of those children who received spam with pornographic content 18% told their
parents about it; younger children were more likely to do so than the older ones.
Of those who had received e-mail which they found unpleasant or which frightened
them, 65% deleted the messages instantly, 29% told a friend, 26% told an adult and
24% tried to prevent further e-mails from the person who had sent the e-mail.
Of those children who had met someone offline that they first met online almost all
said that nothing unpleasant had happened.
Evidence that children accept and respond to parental advice – younger children talk
to parents, older talk to peers
12-14 yrs accept offensive material is part of online experience – they ‘think before
they click’, and practice range of cautious practices
Common strategies among 9-16 yrs – give false info when asked for personal info or
ignore request, ask to be left along, block unknown senders, report to their parents
But also an increase in children who use internet when forbidden by parents or get
around parental rules
Among 7-11 yrs, 21% ask the harasser to leave them alone, 10% stop talking to
strangers, 10% avoid those sites
29% teenagers leave problematic sites/forums, 23% ask harasser to stop, 18%
ignore disturbing messages, 1% are curious and continue the conversation.
1 in 5 of those frightened or hassled online told an adult
Fewer trust info on the internet than in 2003 (down from 49% to 37%)
Fewer willing to publish post info online for safety reasons than in 2003
85% of those who meet online contacts offline would not tell a parent after an
unpleasant meeting
16% stated having received an e-mail that bothered or frightened them (13% in
2003); over half of them deleted it at once, and 1/3 told a friend about it.
1/3 of those (9-16 yrs) who visited a pornographic site say that they didn’t give it
much thought, while more than in 2003 thought it was funny. Those who had seen
such a site generally did nothing about it or did not visit it again. A higher number
than in 2003 stated that they did not tell anyone about having visited such a site.
22% of children who use the Internet have received pornographic junk mail, but very
few have told their parents.
After the visit to a website with violent or gruesome pictures, there is an increase in
the number that said they wished they had never seen it, while somewhat fewer
thought it was cool. [Source: SAFT, 2006]
•
France
•
Greece
•
Iceland
•
•
•
•
•
•
Ireland
•
•
•
•
Italy
•
•
Norway
•
•
•
•
•
•
•
•
45
Poland
•
•
•
•
Most follow safety rules for stranger danger (refuse disclosure of personal info, etc);
yet most also have ignored these rules in last year – 64% gave online contact their
phone number, 42% gave address, 44% gave their photo, and many accept
invitation to meet.
More than 30% of the 12-17 yr olds who had been induced to sexual conversations
felt frightened/ scared about this situation (Wojtasik, 2003).
Many victims of cyber-bullying responded to this cyber-bullying with neutral
emotional reaction: 35% of victims of humiliation/poking, 43% of victims of
threats/blackmailing, more than 80% of those who received unwanted photo or video
and 54% of filmed respondents responded with neutral emotional reaction.
Nearly half (49%) of the participants who have received links to pornographic
websites have used them (29% have done this repeatedly)
Cyberbullying - most tell no-one – very few tell an adult
Most teens do not falsify personal info online or ask online contacts for help; if feel
risky online, they disconnect.
Spain
•
•
Sweden
•
1 in 4 of those who see online porn are disgusted by it
The
Netherlands
•
Online porn associated with recreational attitudes to sex and negative views of
women as sex objects
The UK
•
•
54% of those who saw online porn – not bothered, while 20% were disgusted
Though some go to offline meetings, 66% took a friend and most told a parent or
friend
31% of 12-15 yrs check reliability of websites and 67% trust most of what they find
online
•
Source: Findings reported in National Reports (see www.eukidsonline.net)
The above findings are surely suggestive, though they are insufficient to draw strong conclusions regarding
children’s ability to cope, or otherwise, with the consequences of online risks.
Qualitative research points to a series of strategies that children are developing to cope with online risks.
How these are applied, and whether they are effective, remains unknown, but the manner of reporting
suggests that children feel in control and confident in using these strategies.
Some countries report fairly high awareness of risks, yet this may not reduce risky encounters; in other
words risk awareness does not necessarily translate into risk avoidance. Especially perhaps in countries new
to the internet (e.g. Poland), there is a disconnect between safety awareness and children’s behaviour –
although the gap between safety awareness and safe practices is familiar also in all countries.
Research is sorely needed that follows children from exposure to risk online through to their coping
strategies and, then, the consequences if any. This should include an account of strategies tried, outcomes
and reasons for responding in particular ways, plus associated emotions - immediate or longer-lasting.
These findings suggest that awareness-raising should be continued for all types of risk encountered by
children online, with more attention to how children do and should cope when they encounter such risks.
Many studies report that only a small minority tells an adult, though it appears that children are developing
their own strategies to respond to online risk. Whether these are effective or not remains unknown.
2.2.5
Relation between online opportunities and risks
Research question R2.2.5: What is the relation between online opportunities and risks?
The UK Children Go Online found that there is a positive and high correlation (0.55**) between number of
online opportunities and number of online risks for 12-17 year old internet users. This led the EU Kids Online
researchers to seek similar findings in other countries. Although the Estonian report noted that daily users of
chatrooms and social networking sites are more likely to meet online contacts/strangers offline, no other
countries could produce similar findings. It seems that individual projects examine either opportunities or
risks but rarely both. When they do include both, they tend to analyse the data for each separately, resulting
in a missed opportunity to understand the relation between online benefits and risks.
46
Theoretically, the possibility of a positive correlation – thus far found only in the UK – is important because it
contradicts the notion that as children do more online (becoming more confident or expert or even, older)
they learn to take up more opportunities and avoid more risks. Instead, like learning to ride a bicycle or read
a book, those who take up benefits are often also those who encounter risks, and vice versa – limited
experiences are safe but limited.
Significantly, if the positive association between online opportunities and risks were found elsewhere, this
would strengthen the dilemma posed to policy makers by the UK Children Goes Online project, namely that
increasing opportunities tends to increase risks, while decreasing risks tends to decrease opportunities. No
way has yet been found, it seems, to increase opportunities while decreasing risks.
2.3. Attitudes and skills15
This chapter explores children’s internet skills and attitudes. It examines the relationship between skills and
risks using three variables – age, SES and gender to test four hypotheses:
H2.3.1: As children get older they gain greater online skills, including self-protection skills
H2.3.2: Children who use the internet longer and for more activities develop more skills
H2.3.3: There are inequalities in skills and literacies as a consequence of inequalities in SES
H2.3.4: There are gender differences in the levels of skills (higher for boys)
Hypothesis H2.3.1: As children get older, they gain greater online skills (or Internet literacy, including skills
enabling self-protection from online risk).
Are children’s online skills influenced by age? It is hypothesized that as children get older, they gain greater
online skills (or Internet literacy) and so gain more opportunities and also, presumable, gain the skills
enabling self-protection from online risks.
Several countries had research available that examined this hypothesis, and in each case, it supported the
hypothesis, showing that skills increase with age.
!
For example, in Austria, research suggests that younger children (under 10 years) often have few skills
in dealing with the Internet and consequently they estimate their skills as limited. [102] Older children
and adolescents are more experienced and therefore their use is far more skilled and safe. For the age
group of 12-16 years it seems that safety and competence in dealing with the Internet primarily depends
on the frequency of use; boys are a bit more competent and confident in using the internet. [72].
!
In Portugal, an on-line survey (Cardoso, Espanha, & Lapa, 2007) showed a positive correlation between
age and skills. When asked about who the Internet expert at home is, 81.5% of 16-18 years old
consider themselves to be the ones to have more skills, against 41.7% of 9-12 years that declare the
same. Self-perception of expertise is not the same as having actual skills, but gives us an idea of its
differentiation by age and corroborates other indicators.
!
A French survey (N=468, aged 10-21 years) shows clearly how skills develop with age (Martin, 2004).
Table 2.14: French children’s internet skills by age
Skills (%) by age
Can use a printer
Can install a software
Can surf
Can use a scanner
Can delete web site history
Can maintain the computer
Manage files and directories
11-13 yrs
92
74
73
56
26
26
39
14-15 yrs
98
81
88
64
49
40
50
15
16-17 yrs
99
83
99
75
57
50
65
18-20 yrs
99
85
97
79
47
62
70
This chapter has been written by Helen McQuillan based on comparative analyses done by Yiannis Laouris (H2.3.1,
H2.3.2), Helen McQuillan (H2.3.3), Cédric Fluckiger & Benoit Lelong (H2.3.4).
47
! In the UK Children Go Online project (Livingstone & Bober, 2004), most 9-19 year olds (56%) who use
the internet at least weekly consider themselves ‘average’ in terms of their online skills, though one third
(32%) consider themselves ‘advanced’. Specifically, 9-11 year olds describe themselves as 18%
beginners, 55% average 20% advanced and 7% expert. Among 12-15 year olds, the figures are 4%,
58%, 33% and 4% respectively. Among 16-17 year olds, they are 2%, 52%, 40% and 6%. Overall, age
(for 12-17 yr old internet users) is positively correlated with online skills, r= 0.27** and self-efficacy,
r=0.12** (NB the online skills measure covered a range of skills but did not specifically measure selfprotection skills.
For the most part, it is unclear if increasing skills results in an increasing ability to cope with or avoid online
risk. However, in Cyprus (EC, 2007), developing online skills could be seen to aid safety awareness.
! Among Cypriot boys aged 9-10 years, when respondents were asked as to whether encountering ‘virus’
problems changed the way they use the Internet it was evident that boys do not realize the seriousness of
such a problem. But among boys aged 12-14 years, they had become very cautious in giving away their
personal information on the Internet. They also mentioned that one should be very careful in the way they
spell words on the Internet as misspelling a word might redirect to other irrelevant sites, and that with
antivirus programs installed on their computers, they feel more protected although they mentioned the
downside of these antivirus systems is that they have a license that expires very often/too soon.
! In Estonia too, older children (9 to 14 year olds compared with 6 to 8 year olds) tend to be more aware of
online risks (Turu-uuringute AS, 2006) and they also do “more advanced things“ (e.g. using MSN,
sending and reading emails, downloading music, movies, software and video games) on the Internet
(Mediappro, 2006).
! Similarly in Iceland (Capacent Gallup, 2007), as children grow older they seem to be more cautious
towards the internet. For example, 20% of children aged 9 say they try to verify information obtained from
the internet, compared to 53% of children aged 15. Although it seems to contradict this that the younger
children are less likely to believe that information on the internet is accurate and trustworthy (some 13%
of 9 year olds say that information obtained through the internet can be trusted, compared to 43% of 15
year olds), research is growing to suggest that increased skills brings trust (as the user gains the ability to
discern valuable from misleading information) while novice users are often the most distrustful.
In conclusion, where evidence is available, it does seem that increasing skill may increase self-protection.
But one should not be complacent about these growing skills. The Norwegian report observes that although
many young people have good skills and knowledge about Internet and chat, and what precautions the need
to take when chatting, still, some adolescents “forget” to take the necessary precautions and have
unpleasant experiences on the Net and when meeting new “friends” face to face. After all, the link from
safety knowledge to behaviour change is one of the most uncertain. Encouragingly, the SAFT projects
showed that, from 2003 to 2006, the overall tendency to reveal personal information online fell, as the
population gained in safety awareness).
Although there was no evidence available for this hypothesis in Belgium, Bulgaria, Czech Republic,
Denmark, Germany, Greece, Ireland, Poland, Slovenia, Spain and The Netherlands, there is no reason to
suppose that gaining such evidence would contradict the hypothesis. It seems that, as children grow older
their level of skill increases, and this is likely to include their abilities to protect themselves from online risks.
Hypothesis H2.3.2: Children who use the internet longer and for more activities develop more skills.
Although it would seem obvious that children who use the internet for longer and for more activities would
develop more internet-related skills and literacies, only three countries provided data for this hypothesis. In
Austria, Norway and UK, findings showed a positive correlation between frequency of use and online
competence and safety. However, with such few data, generalisations are unsafe.
Hypothesis H2.3.3: There are gender differences in the levels of skills (higher for boys).
In studies where children/young people self-report and self-evaluate their internet skills boys tend to rate
themselves higher than girls. There is little evidence of tested or demonstrated skills levels. Comparable
48
research in Italy and Poland (EC, 2007; Eurydice, 2004) reports higher skills levels for boys in four ICT
activities: downloading files, using PowerPoint, creating web pages and sending email attachments.
Conversely, in the UK and Portugal fewer boys than girls rate their internet skills poorly. Girls are more
confident than boys in their information-searching skills. In the UK girls aged 12-15 who use the internet at
home are significantly more confident about using the internet than the same age boys (97% vs. 91%). Also,
more boys than girls report that they cannot find what they are looking for on the internet. It is unclear
whether this confidence leads to greater safety awareness or practice. Possibly too, girls’ preferences for
communication and information searching are seen as lower-level skills, though technical skills have long
been attributed to boys.
In Italy, two studies report higher level digital skills for boys. Accorsi and Gui’s (Accorsi & Gui, 2006) study
reported that a significantly higher percentage of boys than girls (56% vs 35%) demonstrated high level
digital skills. Similarly in France, more than twice as many boys (57%) than girls (26%) consider themselves
as the most skilled member of their household. In the UK, more boys (35%) than girls (28%) consider
themselves as ‘advanced’ internet users. They are also significantly more likely than girls to be aware that
there are illegal as well as legal ways to access films, music and software on the internet (80% compared to
72% of girls).
Boys tend to describe themselves as more expert and claim to have more technical and advanced skills.
Research from Norway reports that boys are more technically-focused than girls in their online activities.
Research from the Netherlands also examines mastery of skills. While there are no gender differences
reported in information sharing and word processing skills a lot more teenage boys than girls say they are
skilled in activities such as installing anti-virus programmes, upgrading software or replacing a hard drive.
Austrian research links boys’ higher skills levels with their more frequent use of the internet. In Sweden boys’
greater technical skills are considered to result from their use of a wider range of applications including
downloading, skype and web development. A similar level of skill is not apparent in their self-protection or
safety skills.
Different internet activities result in different ranges of skills. Evidence suggests that boys and girls may be
developing different skill sets based on their different activities, but it is inappropriate to nominate more
technical/machine focused skills as higher level than communication and information literacy skills.
Research from Bulgaria reports that boys boast of being more knowledgeable, more skilled and more daring
than girls. Kirwil’s Polish study shows how this can have consequences for boy’s greater exposure to
potential risks. Young teenage boys (13-15) report knowing how to deal with internet filters blocking access
to pornographic websites and also how to set them up again after disabling them (Kirwil, 2002).
Caution needs to be expressed in drawing conclusions for several reasons. Boys and girls are no
homogeneous groups and there are many differences in skills among genders as between them. Where
evidence is available it is difficult to compare, there are inconsistencies in skills measurements, and much
data refers to self-perceptions of skills levels rather than tested or demonstrated skills.
The different interests of boys and girls, as well as social norms influence their choices of internet activities,
resulting in different skills sets and exposure to risks. This should be considered in awareness and safety
campaigns
Hypothesis H2.3.4: There are inequalities in skills and literacies as a consequence of inequalities in SES.
This hypothesis sought to examine the relationship between inequalities in skills and literacies and
inequalities in socioeconomic status (household income, parental education and social class). This has not
been examined in depth in any study. Limited evidence is provided by four countries: France; Italy; the
Netherlands; UK. In each case, the research had a slightly different focus. In higher SES households in
France parents are more familiar with computers and therefore able to help their children. In lower SES
households friends play a greater role. There is no indication of the impact of this help on either skills or
safety awareness or strategies.
In Italy there is a close connection between parents’ professional status and the level of children’s digital
skills. In the Netherlands the important variable is the young people’s education level. Teenagers with higher
education levels have more ICT skills.
49
Home ownership of computers appears to be an important variable for skills development. UK research with
regular internet users aged 12-17 shows a correlation between SES, online skills and self-efficacy. However,
this was not the case across SES in children who have similar levels of home internet access. The Ofcom
Media Literacy Audit (Ofcom, 2006) found that children aged 12-15 years from minority ethnic groups are
less aware of film, music and software illegal downloads (65% compared to 76% across the UK).
Conclusion
The relationship between online skills and risks cannot be conclusively addressed based on the comparative
research examined here. Even establishing the links between age, gender, SES and online skills is tentative
without considering some other variables.
Examining internet skills is problematic for a variety of reasons. Research on internet skills is sparse and
data exploring links between skills and risks even more so. Skills are poorly conceptualised and measured in
internet research, with little definition of task-based, technical, internet literacy or self-protection skills.
Research tends to focus on skills in using internet applications, with very little examination of content
creation or more creative and collaborative use of the internet.
The context of skills acquisition has a bearing on attitudes, self-confidence, risk-taking behaviours, selfprotection strategies as well as skills. The Eurobarometer study (EC, 2007) notes the importance of informal
learning for skills development. Learning to use the internet is described as a process that is easy and quick.
Taught the basics at home by parents they develop their skills through self-learning and peer observation
and support. Comments such as ‘there is nothing to learn’ when it comes to the internet suggest that
attitudes to learning and skills need to be considered more carefully in future research and internet
awareness and safety campaigns.
2.4. Mediation by parents, teachers and peers16
This sub-chapter will summarize empirical evidence from the European countries on how parents, teachers
and peers mediate children’s internet use. Unfortunately, very little is known about the actual influence of
teachers and peers, although it might be assumed that they play a crucial role. Therefore, the following
findings mainly refer to parents’ behaviours.
Research question R2.4.1: To what extent do parents set rules for different media?
In the Eurobarometer survey parents were asked whether they have set rules about using television, mobile
phones, games consoles and the internet. The results shown in figure 2.4.1 are based only on children who
use the internet.
!
It was expected that more rules would be set for younger children. However, the general pattern of
results is an inverted U-curve. Until the age of 10 years old, there is an increase in setting rules; after
that there is a strong decrease.
!
The shape of the curves differs substantially across media. Until the age of 12-13 years old, television is
the most regulated medium.
!
Since the results in figure 2.6 are based on internet users only, this suggests that parents of children
who use the internet are more likely to set rules for television than for the internet. After the age of 12/13
years more rules are set for internet than for television.
!
Other media are less regulated than either television or the internet. The results reflect the strong
position of games among younger children, whereas rules for mobile phones are becoming more and
more important for teenagers. Parents of children aged 14 years and older are most likely to set rules
for mobile phones.
16
This chapter has been written by Thomas Wold based on comparative analyses done by Uwe Hasebrink (R2.4.1),
Thomas Wold (R2.4.1, H2.4.1), Helen McQuillan (R2.4.3), Cédric Fluckiger & Benoit Lelong (R2.4.2), and Lucyna Kirwil
(R2.4.4, H2.4.2).
50
!
There are almost no gender differences with regard to setting rules about using television and the
internet; the curves for boys and girls are very similar. The only exception is that parents set
substantially more rules about playing games for boys of all age groups. In the face of boys’ particular
interest in games, this finding is highly plausible.
Figure 2.6: Parents who have set rules for using different media
(in % of parents, whose children use the internet; EU 25)
90,0
TV
Mobile
Games
Internet
Any rules
80,0
70,0
60,0
50,0
40,0
30,0
20,0
10,0
0,0
0-5 yrs
6-7 yrs
8-9 yrs
10-11 yrs
12-13 yrs
14-15 yrs
16-17 yrs
less than 18 years.
If we look at different countries, noticeable differences can be observed. Parents in the different countries
differ with respect to their tendency to set rules about their children’s media use (see table 2.15, column 1).
Whereas almost 80 per cent of the Irish parents have set rules, this is the case for only around 50 per cent of
parents in Slovenia. Although the calculations are based only on children who use the Internet, parents are
still more likely to set rules about television viewing than about using the internet.
Again, there are clear differences between the countries with regard to the medium, which is more or less
regulated: column 4 of table 2.4.1 shows the difference between the percentage of rules for television and
rules for the internet.
The emerging pattern is quite clear: parents in Sweden, Denmark, the Netherlands and Estonia are more
likely to set rules for the internet, whereas Portugal and Poland pay more attention to regulating their
children’s television behaviour.
The ongoing diffusion of the internet seems to raise parents’ internet literacy and awareness of risks; thus
they increase their efforts to regulate their children’s internet use.
Table 2.15: Parental mediation (I): Rules set about children’s media use
(in per cent of parents, whose children use the internet)
1)
at least one
Rules set for …
2)
TV
51
3)
Internet
4)
Difference
EU 25
Ireland
Netherlands
Spain
France
Sweden
Austria
United Kingdom
Belgium
Germany
Poland
Italy
Estonia
Czech Republic
Portugal
Greece
Denmark
Cyprus
Slovenia
medium
67.9
79.7
74.9
74.5
74.2
72.8
70.1
68.7
68.1
65.6
64.8
64.2
62.6
60.3
57.0
56.9
55.9
54.8
50.6
42.3
60.5
44.8
49.7
52.4
34.8
42.6
45.3
45.9
45.3
36.0
31.9
23.9
23.3
39.0
35.5
23.8
37.3
30.3
37.6
61.8
53.4
40.1
41.9
58.2
35.1
41.8
44.6
42.4
24.2
23.5
32.2
24.2
22.3
31.2
35.5
34.8
25.3
column 2)-3)
4.7
-1.31
-8.55
9.64
10.46
-23.41
7.53
3.51
1.32
2.89
11.78
8.36
-8.36
-0.95
16.66
4.29
-11.72
2.54
5.01
less than 18 years.
Research question R2.4.2: To what extent do parents make use of filtering or blocking tools?
One option for parents to regulate their children’s internet use is the use of filtering or blocking systems,
which can help to avoid the child coming into contact with potentially harmful content. On the European level,
28 per cent of parents say that they use these filters at home (see table 2.16); even more (31%) say so for
schools.
Again, there is considerable variation across countries, with UK parents and UK schools (as perceived by
parents) being the most frequent users of these technical tools; particularly the result for the UK schools is
notable. At the other end of the spectrum are Portugal, Estonia, Bulgaria and Slovenia, where filtering tools
are rarely used.
It is also important to consider the extent to which parents are aware of filtering programs and how they can
be used. Surprisingly enough, on the European level the percentage of parents who do not know about
filtering tools is less then 10 per cent. Awareness is lowest in the Czech Republic and Poland – note that
these countries were classified as being within the highest risk category (see chapter 2.2).
52
Table 2.16: Parental mediation (II): Use of filtering/blocking tools
(in per cent of parents, whose children use the internet)
EU 25
United Kingdom
Ireland
Germany
Netherlands
Spain
France
Austria
Cyprus
Belgium
Italy
Sweden
Poland
Denmark
Greece
Czech Republic
Portugal
Estonia
Bulgaria
Slovenia
Filtering/blocking tools …
at home
at school
unknown
(%)
(%)
(%)
27,6
30,8
8,0
46,2
71,0
4,2
35,2
43,4
2,8
29,8
21,5
5,0
28,0
31,6
0,8
25,6
14,4
4,4
25,4
26,6
10,2
21,8
17,7
9,5
20,7
20,7
0,0
20,6
9,8
6,7
20,2
10,6
6,7
19,9
26,5
0,9
18,8
30,0
21,3
18,4
20,3
1,9
11,9
22,0
13,6
10,1
20,7
21,8
8,7
20,7
10,9
7,0
11,3
10,2
6,6
8,8
7,7
5,1
1,9
17,2
Source: Eurobarometer 64.4 – Special No. 250: Safer Internet, December 2005;
Research question R2.4.3: What are the main strategies of parental mediation practised?
Evidence on this question is available from 17 countries, though there is great variation in the evidence
available from the different countries. There seems to be a clear research gap here.
The different strategies of parental mediation are, in order of importance:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Time restriction; mentioned by 11 of 17 countries.
Supervise/control; mentioned by 8 countries.
Talk to/teach children about safe usage; mentioned by 8 countries.
Filtering software; mentioned by 7 countries.
Rules against revealing personal information; mentioned by 6 countries.
Not to visit certain sites; mentioned by 6 countries.
Monitor visited web pages/check history file; mentioned by 5 countries.
Rules against meeting someone they have only met online; 4 countries.
Not talking to strangers in chat rooms; mentioned by 4 countries.
Rules against downloading files; mentioned by 3 countries
Not allowed to buy things, mentioned by 2 countries.
Rules against foul language/bad behaviour; mentioned by 1 country.
The material seems to be somewhat inconclusive; it does not give us sufficient evidence to claim that the
various strategies for parental mediation do not exist in the countries that have not mentioned them, but it
does point to gaps in the available research material.
1. Time restrictions
Imposing time restriction is the most common strategy for parental mediation. Eleven of 17 countries have
material confirming this. In nine of these countries it is also cited as the most important means of mediation.
There might be different reasons why time restrictions are so widespread. To divide internet access between
family members and to make sure school is given priority are two considerations that are clearly linked to
time restrictions.
53
Economic motives might also affect this depending on how the Internet use is charged; if they pay a fixed
sum every month, as is usual with broadband, there is no economic motivation for time restrictions. But if
they pay for every minute online and if they have access via a modem that blocks the telephone line, then
there are economic and infrastructural reasons for time restrictions.
Time restriction is also an easy strategy for parental mediation. It is easy to decide fixed rules for time spent
with computers or games, and to use it as a reward or punishment. Time restrictions on internet use are also
in harmony with a general attitude that Internet should not take up too much time and that the children need
variation in their activities. For further research it would be interesting to ask parents why they have time
restrictions, how they organize them and how they handle them, and to ask children what they think about
the rules and if they have strategies for bending these rules.
2. Supervision/control
This form of parental mediation is mentioned by eight countries, and cited as an important strategy in seven
of them. It is done either by sitting next to the children while they are online, or by watching the
screen/checking up on them from time to time, preferably with the computer in a shared room. Will this
strategy lessen in importance or become more difficult with the development of the bedroom culture, where
more children and young people have a computer in their bedroom?
3. Talk to/teach children about safe usage
8 countries have findings suggesting that parents talk to their children about Internet usage and try to teach
them about safe usage. A large majority of the parents in the 8 countries do this often or from time to time.
However, the numbers from the Czech Republic indicate a possible source of error: 82 % of the parents
state that they talk to their children about safe internet usage, but only 39 % of the children state the same.
The discrepancy is quite large. Are the parents over-reporting as part of being “the good parent”, or are the
children under-reporting as part of becoming autonomous? Or do they understand the question and interpret
the conversations differently? These are possible questions for further investigation.
4. Filtering software
In 7 of the countries, the parents state that they have installed filtering software, although it is usually a
minority of the parents: 43 % of the parents in Germany, 33 % of the parents in the UK, 32 % of the parents
in France, 21 % of the parents in the Netherlands, 11 % in Poland.
The Spanish report displays strongly divergent numbers. According to one study, 44 % of the Spanish
parents have installed filtering software, but a different study claims the number to be 11 %. A possible
reason for the divergence can be that some parents might confuse filtering software with firewalls, anti-virus
and anti-spamming software, and this might not be reflected in the study. On the other hand, it is also
possible that filtering software come as a part of a package without the parents being aware of it.
Further research on this question ought to examine whether the parents know what a filter is, how they use it
and what purpose they want the filters to fulfil. Information about this is sparse, but in the Spanish report the
filters are mostly related to pornography, terrorism or violence rather than the use of chat-room, instant
messaging and e-mail.
5. Rules against revealing personal information
6 countries have mentioned this, and it is common in five of them (Iceland 63 %, Italy 69 %, the Netherlands
57 %, also common in Ireland and Spain). In Poland, only 4 % of the parents state that they have such rules.
Why is this not more common? Will we see a change here as themes like identity theft are hitting the
agenda, or will the urge to (and the right to) publicise oneself remain a priority?
6. Rules not to visit certain sites
This is mentioned by 6 countries, but only in Spain (69 %) and Italy (72 %) do we have high numbers. It also
seems to be common in Ireland and Belgium, but no percentage is stated. In Poland (9 %) and Austria only a
minority have such rules. The material does not say much about what kind of web sites are prohibited, but
Belgium and Italy state that parents have explicit rules against sex or “dirty” content. Why do so few have
such rules? Is it because they trust that their children will not visit such pages anyhow, do they trust that a
filter makes such rules unnecessary, or do they feel that they cannot control this aspect of their children’s
media use anyway?
54
7. Monitoring visited web pages/checking the history file
This is mentioned by only five countries, and only a minority of the parents state they have checked their
children’s Internet history. The Netherlands (38%) and the UK (30%) have the highest numbers in this
category. The other three countries are Spain, Ireland and Austria.
There is an important distinction between supervising and monitoring the children’s Internet use. Supervising
means open observation and discussion or being in the same room while the child is online, and is quite
common, while monitoring means that the parents are secretly checking which pages the child has visited.
Very few parents perform this technical equivalent of going through their children’s drawers; is it out of
consideration to the children’s right to privacy, or is it simply because the parents lack the technical
competence to do this?
8 and 9: Rules against meeting someone they have only met online, and not talking to strangers in chat
rooms
These related mediation strategies are given high priority in four countries (Iceland, Ireland, Norway and the
Netherlands have rules against meeting someone, while Iceland, Ireland, Norway and Italy have rules
against talking to strangers in chat rooms). Unfortunately, evidence is lacking (though rules may not be) in
several countries. It is also a possibility that parents have general rules concerning who their children are
allowed to talk to and socialize with, hence eliminating the need for special rules concerning chat rooms.
“Don’t talk to strangers” is a quite general rule, applicable both for the playground and the Internet.
10., 11., and 12: Downloading files, buying things, use of foul language
The existence of rules against downloading files is mentioned by three countries; Belgium (more common for
girls than boys), Estonia (less than 10%) and the UK (17%). In other words this is not a very widespread rule.
Rules against buying things on the Internet in only mentioned by a small percentage in two countries, and
only Norway has mentioned rules against foul language and bad behaviour, although in Norway it is quite
common. The considerations for this might be the same as for rules against meeting strangers: It can be
more relevant in some countries than others, depending on the diffusion of Internet and the way it is being
used, and it can be that more general rules concerning language, behaviour and money eliminate the need
for special Internet rules.
Wish for guidance
It is quite clear that the parents wish more guidance concerning their children’s use of online media. The
parents have suggested different ways of doing this. In Austria the parents suggested arranging special webportals for children with a child safety lock offered by internet providers that would preclude complicated filter
software. French parents thought an information guide would be useful, while parents in Norway and Spain
said they needed more information on safe use of electronic media. The Spanish parents particularly wanted
more information about how to protect their children from illegal or harmful contents, and 38% of them did
not know where to report illegal Internet content. They said that the principal means of receiving this
information should be from school (55%) and from the media (32%).
In future research, several kinds of country classification would be possible:
!
Preference for restrictive versus flexible or active regulation (e.g. forbidding or talking)
!
Preference for time restriction versus content supervision (as both are familiar to parents as regards
television regulation)
!
Human interaction versus delegation to technical systems (e.g. filtering and monitoring software)
At present, however, the measures are too variable across countries to permit reliable and comparable
analysis or country classifications.
Research question R2.4.4: Are there SES differences in parental mediation?
Parental mediation has been very little studied throughout Europe. However, evidence could be found in
several countries, that all show differences in parental mediation related to SES differences. Generally, it
seems clear that there is more parental mediation in higher SES families, as reported by parents.
55
This parental mediation may be described differently from one country to another, but in every country, upper
class or highly educated parents claim to implement more parental mediation than other families.
However, with the exception of Ireland, results are not very accurate (c.f. National Reports). French upper
class parents are more likely to set up time restrictions and inculcate self-limitation (Pasquier, 2005).
Icelandic children whose parents have a higher level of education are more likely to say that their parents
check which websites they visit and check on them while they are browsing the Internet. Spanish parents
with high social status are more likely to supervise their children than parents in the middle social status
group (Tezanos, 2006), while in UK higher SES parents implement more rules and practices (Livingstone &
Helsper, in press). German working class parents are said to be less interested in the media consumption of
their children. In the Irish report there is evidence that the mother is more likely than the father to supervise
the children’s media use, and in Ireland children are more likely to talk to mothers than fathers about their
online activities (Webwise, 2006) but apart from that we have little information on the differences between
maternal and paternal mediation.
Future research should use clear, comparable measures to study parental mediation, including the use of
parental control tools. Most importantly, research is needed that seeks to evaluate the effectiveness of these
different forms of parental mediation on the risks and opportunities experienced by children. There are some
unexpected results that are still to be explained, e.g. on higher SES in Spain and minority ethnic groups in
UK.
Future research should also try to explain SES differences, rather than simply describing them. Firstly,
control of the child’s relationships appears to be stronger among higher SES families than it is among the
working classes. Sociological research has shown these parental behaviours depend on different ways to
secure social capital and socially homogeneous relation networks for the children. This could explain why
“chatting” with strangers on the Internet is less accepted by higher SES families, and disappears more
frequently and earlier as the child grows older compared to lower SES families. Secondly, parents with
higher qualifications tend to supervise access to media content more strictly. Their educational strategies are
based on hierarchies that put classical culture (books and “serious” newspapers) above the TV,
entertainment-based magazines and information available on the Internet. Here again, family strategies for
the transmission of cultural resources depend on social and economic status: more research should be
undertaken on the effects of these strategies on parental control of internet use.
Research question R2.4.5: Are there gender differences in parental mediation?
Comparative analysis of this research question is difficult. Gender differences are reported in three countries:
Iceland, Ireland and Spain. Two countries reported no gender difference in parental mediation: Belgium and
the UK. Limited evidence is available from seven countries: Estonia, France, Germany, Greece, Poland and
Portugal. There is no evidence available in the remaining 10 countries.
Estonian findings (Metsoja, 2006) suggest that girls tend to talk more about the internet with their parents
than boys do, but the difference in not very large; 12 % vs. 7 %. In Greece, slightly more girls (67%) than
boys (64 %) are expected to phone home during the day, but at the same time, fewer girls (27%) than boys
(29%) feel that they are under parents’ surveillance, and more boys (54%) than girls (41%) report that their
mobile phone use is regulated. In Iceland there is a significant tendency that girls are more subject to
parental control; they have more rules regarding Internet use than do boys (94 % vs. 89 %) and 33 % of the
girls and 23 % of the boys have parents sitting with them while they use the internet. In Ireland there are also
different rules for girls and boys. Details are missing here, but it is clear that mothers are more protective of
girls, and that the rules predominantly relate to “stranger danger”. In Spain there are also more rules for girls
than for boys, although the differences are rather small except for one rule: 63 % of the girls are not allowed
to give out personal information online compared to 42 % of the boys.
It is not possible to determine systematic cross-national patters or variations from the limited data. Girls
seem to be subject to more parental supervision than boys, but from the available material we cannot decide
if this results in less risk taking behaviour. It seems that parents think that girls are more vulnerable and in
need of protection than boys, but Norwegian findings clearly indicates that more teenage boys than girls
have experienced unwanted sexual attention online (SAFT, 2006).
More attention should be given to parental perceptions of risk, exposure to risk and risk-taking behaviour
among boys and girls. Parents tend to talk more to girls about internet dangers and safety than boys. Safer
56
internet campaigns could highlight the need for better communication with boys. As solitary and bedroom
use of the internet grows, and supervision and knowledge of children’s internet use diminish, it is likely that
these will be used as opportunities for more risk-taking behaviour.
More research should also be done on gendered forms of parental control as a whole (i.e. not only control of
internet uses, but also control of clothes, school performances, practices of going out, etc.) For example, the
presence of girls in public spaces (such as streets or bars) is generally more controlled than boys'? These
attitudes depend on family strategies which differ from one cultural area to the other: this approach could
thus be fruitful for cross-national comparisons.
Hypotheses H2.4.1: As children grow into teenagers they are subject to reduced parental mediation in their
use of the internet.
We found evidence to support this hypothesis in 11 countries: Austria, Estonia, France, Greece, Iceland,
Ireland, Italy, Norway, Spain, Sweden and the UK. We found evidence to contradict in 2 countries: Poland
and Portugal.
The answer to the hypothesis would be: it depends on the age. The Eurobarometer dataset provides some
results on how parental mediation evolves while children are growing into teenagers (see above). As already
seen above for the general rules about using different media, answers provided by parents from EU 25 are
quite clear. From 6 to 12, while children are growing into pre-teenagers, European parents tend to make
more rules. After 12, the use of rule decreases.
Figure 2.7: Rules set for the use of the internet by age
(% of children who use the internet)
He\she is not allowed to give out any personal information
70
There are some websites that he\she is not allowed to visit
He\she is to tell me\us if he\she finds something on the Internet that makes
him\her feel uncomfortable
He\she is not allowed to use rude language in e-mails or chatrooms
60
50
He\she is not allowed to meet in person someone He\she only met on the
Internet
He\she is not allowed to copy documents\pictures
40
He\she is not allowed to go to chatrooms\ to talk to strangers in chatrooms
30
He\she is not allowed to play games online
20
He\she is not allowed to do online shopping
10
He\she is not allowed to download music or films
He\she is not allowed to download software
0
17
13
6
9
Rules regarding how much time he\she is allowed to spend on the Internet
Keeping phone lines free at certain times of the day
to
14
to
10
to
6
U
nd
er
Source: Eurobarometer 250, ibid.
Ensuring that access to the Internet is shared fairly between family
members
Parental rules reflect the phases of the child’s growing independence. For younger users, parental
constraints are rather low. An explanation would be that younger children tend to follow parental
expectations, making explicit rules unnecessary. As children grow older, they tend to experiment beyond
what parents allow them to do. They want to explore the world, to learn about youth culture, to meet peers
outside there familial environment. Parents feel the need for specific rules and parental control tools.
When teenagers are growing older, their growing independence is expressed through increased freedom,
and parental control tends to loosen. Almost all national reports reflect this tendency. Countries providing
results focused on younger children show that parental mediation is growing with the age, while countries
dealing with teenagers agree that parental mediation is decreasing with age.
57
Evidence to support
Where national reports are mainly focused on teenagers (for reasons of research availability), the general
picture is very clear: Parents impose fewer and less strict rules concerning media usage as their children
grow into teenagers. Actually, the development starts earlier than that, and there is a time continuum where
rules decrease with age, although there are often small, but significant, differences between the age groups.
For instance, in Estonia 12 % of children aged 12-13 say that their parents have forbid them to visit certain
web sites, while 9 % of 14-16 year-olds claim the same, and 1 % of 17-18 year olds. The same goes for
downloading music or movies (14 %, 8 % and 4 % respectively) and time restrictions (18%, 19% and 8 %)
(Turu-uuringute AS, 2006).
Time restrictions: In Italy 45% of parents limit the total amount of time spent online by their children aged 711, against the 26% of parents of teenagers aged 12-19 (Eurispes/Telefono Azzurro, 2007). The Estonian
report also states that time restrictions on Internet use decrease with age. This goes for online games and
other computer games as well, although here time restrictions are still quite common (Mediappro, 2006).
Monitoring visited sites: Austrian findings indicate that parents stop monitoring visited websites as their
children grow older, and in Iceland 78% of the 9 year olds say that their parents check which web sites they
have visited compared to 22% of the 15 year olds (www.eukidsonline.net). The French report states that the
proportion of teenagers saying their parents let them do what they want on the Internet increases with age:
63% of 11-12 year-olds, 70% of 13-14 and 78% of 15-16 year olds (Metton, 2006).
Filtering software: There is less use of blocking software as the children grow older. In France, 31 % of 12-14
year olds say they have filters on their computer, while 25 % of 15-17 years old say the same (Martin, 2008).
In Iceland 33% of the 9 year olds say that their parents use devices of some kind to prevent them from
looking at particular websites compared to 9% of the 15 year olds.
Supervise Internet use: In the Icelandic report, 41% of children aged 9 say that their parents sometimes sit
together with them while they surf the internet compared to 13% of 15 year olds (www.eukidsonline.net). In
Ireland it is common that parents of 9-10 year olds sit with their children or are nearby when they are using
the internet, but they rarely sit with older children (EC, 2007).
Evidence to contradict
Poland and Portugal gave evidence to contradict the hypothesis, although the findings are ambiguous. Part
of this can be explained by the age group targeted by the studies (i.e. including younger children). In Poland,
parents with teenagers more often set rules for internet use (27%) than parents having a child at lower
school age (11 %). The older the child, the more rules the parents set parents on the internet use. At the
same time, when is comes to computer use the numbers indicate the opposite: 27 % of parents with children
aged 6-11 set rules for computer use, while 17 % of parents with children aged 12-17 do the same
(www.eukidsonline.net). The reasons for this might be by more intensive usage of the internet by older
children and the growing costs of this activity. Many households in Poland only have access to a very slow
and expensive internet, while offline computer use is free of charge and not prompting restrictions out of
economical considerations.
In Portugal parents tend to have more concern with older children than with younger ones. 43 % of 16-18
year olds claim to have had discussions with their parents about the amount of time spent online, against
36% of 9-12 years olds (Cardoso et al., 2007). This probably has to do with the fact that, according to the
same study, older children use the internet more frequently. Parents seem to be more concerned with the
period of the day that younger children use the internet than with the amount of time: 20% of 8-12 years old
declare to had discussions with their parents about the period of the day they use the internet, while this
figure is slightly lower (17%) in what concerns teenagers (16-18 years old).
It seems to be evident that as teenagers grow older, there are fewer rules and less parental control on all
their activities. It is not surprising that this goes for media use as well. For instance, in Greece, 75 % of 11-15
years olds are expected to call a parent during the day compared to 60% of the older teenagers (Tsaliki,
2008). The Greek report also states that parental economic supervision drops to lower levels as their
children grow up (28% of those aged between 16-19 say that their parents check their bills, 43 % for those
with younger children).
Another interesting finding from the Swedish report is that when looking at the parent’s answers, parental
mediation is very little reduced between the age groups 9-12 and 13-16, but when looking at the children’s
58
answers, the differences are bigger (Medieradet, 2006). This confirms other results showing that parents
seem to overestimate parental control, while teenagers seem to underestimate it (see UK Children Go
Online project). Teenagers, more than children, may want to be seen as more mature and emancipated than
they really are, so that the differences are bigger with teenagers than with 9-12 year olds.
Bedroom culture
In the Norwegian report there is a clear tendency that when children grow older, they get more freedom to
use the internet on their own, and a growing number of children also have internet access in their bedrooms.
The emergence of bedroom culture can make parental mediation more difficult, especially when it comes to
supervising the children’s internet use by sitting down with them or checking up on them from time to time.
The autonomy of teenagers must also be taken into consideration here. The report from Ireland states that
teenagers tend to keep their use of the internet private from their parents. Virtual environments are relatively
free of parental control. Teenagers’ increased use of the internet in their social lives results in a reluctance to
alert parents to its negative aspects for fear of having access blocked by protective parents (Webwise,
2006).
Does increased skill make the Internet safer or more dangerous?
As children grow older, they usually become more competent in their use of the internet. On the one hand,
this competence can be seen as making internet use safer. On the other hand, competence opens up new
opportunities and new risks. Do the parents perceive their children’s competence as opening new risks or
new opportunities?
In the French report, a survey shows that children’s skills grow (logically) with age, and that children with
higher technical skills are subject to a reduced parental mediation (Martin, 2004). Fluckiger’s qualitative
study between 2004-2006 (Fluckiger, 2007) points out how greater skills are needed throughout the
independence and emancipation process, in order for the children (12-16) to build a personal digital territory,
i.e. protecting their instant messaging account with passwords, deleting information about visited web pages,
etc. Greater skills may also allow children to hide some of their uses from their parents (such as deleting
visited websites in the browser’s history). On the other hand, the IFOP study (IFOP, 2006) shows that
parents of older teenagers think their child is exposed to a greater amount of online unsuitable content: 41%
of high school pupils parents think their child has already been exposed to sexual material on the internet
(only 12% in primary school), and 37% think the child has been exposed to violent content (only 8% in
primary school).
Hypothesis H2.4.2: More parental mediation results in reduced exposure of children to online risks.
There is a very little evidence to support or contradict Hypothesis H2-4-1 because 17 countries out of 21
reported that there is no pertinent evidence available. The remaining countries reported conflicting findings.
Ireland found evidence to support the hypothesis that more parental mediation results in reduced exposure
of children to online risks, while Poland found evidence to contradict it. The United Kingdom found evidence
both to support and contradict the hypothesis. Two countries commented on a general basis: Cyprus`
comment supports the hypothesis while Spain’s comments contradict it.
Research question R2.4.6: Is there evidence that particular parental strategies or styles of mediation
effectively reduce the risk that their children experience online?
Four countries replied to this research question. Three countries cited empirical findings suggesting that
some techniques mediating the internet for children might be efficient in reducing child’s online risks.
Belgium:
! Parental control
!
Instructive mediation better than restrictive mediation
Ireland:
! Parental supervision
!
Talking
!
Rules setting
59
Poland:
! Blocking/filtering websites in public sphere, not at home
UK:
! Restrictive mediation indirectly through decreasing child’s activities online
Two findings seem to be interesting and demand further studies:
The UK’s findings (Livingstone & Bober, 2004) on restrictive mediation shows that this kind of mediation may
reduce the child’s risks online indirectly through a reduction in child’s online activities, which should result in
a decrease in the probability of being exposed to a risky situation. A question arises here, as to whether
restrictive mediation might block access to informative and useful information, and might even be in conflict
with children’s right to seek information. In Poland blocking/filtering the websites appeared efficient with
regard to computers located in public areas (Internet or cyber cafés and schools).
Conclusions
Some kinds of mediation are effective and some are not. Findings from Poland that mediation through
blocking/filtering the websites is not efficient at home but efficient in public sphere (i.e. in the Internet café
and school) suggest that social factors may moderate efficiency of at least some kinds of parental mediation.
The selection of specific kinds of parental mediation of the Internet chould result from parents’ aspirations for
the child. Do they want mainly protect a child against frustration, “socially bad things”, social conflicts or do
they want their child to develop self-directness, need for freedom and skills useful for his/her future career?
Parents’ aspirations could determine the type of parental mediation. And parents’ aspirations are themselves
determined by individualistic-collectivistic values orientation to a great extent.
For instance, on the basis of theory of individualism-collectivism (Basabe, 2005; Lee, 2005; Inglehart &
Baker, 2000) people’s attitudes to the internet vary significantly. In individualistic cultures people behave selfcentrally and engage in open interpersonal emotional expression in order to attain their personal well-being.
They have a need for autonomy, independence and individuality. On the other hand, in collective cultures,
group membership occupies a central place and the importance of oneself is only peripheral. People in
collectivist cultures restrain their personal emotions, maintaining their positive relationships through
obedience and unselfishness. This distinction allows one to predict that parents from collectivistic countries
will mediate the internet to their children in more interactive ways than parents in individualistic countries
(Kirwill, 2008). Thus it might be assumed that:
!
Hypothesis 1: Parents from individualistic countries as compared to parents from collectivistic countries
prefer to use mediation allowing a child more autonomy and self-directness online, for instance, setting
instructive rules, and not banning some activities.
!
Hypothesis 2: Parents from collectivistic countries as compared to parents from individualistic countries
prefer to use mediation assuming obedience and respect for parents’ values and rules, for instance,
using restrictive mediation, blocking, imposing time restrictions and banning activities, not using
instructive rules.
Conclusions regarding future research
There is need for research on the types of parental mediation of children’s use of the internet. It can be
instructive or assisting, it can be a kind of flexible monitoring or rigid control, it can be fun or a duty for the
parent. Monitoring children’s activity online is psychologically different from co-viewing TV. New specific
research on the mediation of the internet is needed to investigate the efficiency of various types of parental
mediation and the factors that make specific kinds of parental mediation efficient or inefficient, for instance:
gender, age, parental literacy (on the computer and the Internet), parental attitudes to new technologies or
parental values.
This kind of research should include not only survey studies. Qualitative in-depth studies are also needed to
investigate how parents and children understand and evaluate the different types of mediation.
Promotion of a Safer Internet
Parental techniques for socializing children vary across Europe. Parental techniques of mediation will also
vary across European countries, and it is also possible that the efficiency of the different mediation strategies
60
will differ from country to country. Parents may be more efficient in assisting children in how to avoid online
risk and use online opportunities using a certain kind of mediation of the internet that is consistent with the
values of the culture to which their country belongs.
17
The European Values Survey showed that in 2000 there were four groups of countries in Europe as
described in terms of the values orientation characteristic of their culture: Protestant Europe, Catholic
Europe, English speaking Europe and Ex-communist Europe.
It might be assumed that the way that parents ( or the state, or the school, or the child himself/herself take
responsibility for the children’s safer Internet and parental techniques mediating the Internet may vary across
the countries belonging to different groups defined by individualistic-collectivistic value orientations. In
addition, finding the cases of efficient parental mediation of the internet at the country level may allow us to
describe the conditions under which the technique is efficient and to disseminate knowledge concerning how
it should be put into practice in other countries.
A suggestion is that research on a country level is needed to establish the conditions under which some
kinds of parental mediation are more efficient than others. And then at the level of groups of countries we
need research on the efficiency of different ways of implementing the recommended techniques.
2.5 Conclusions
2.5.1
Conclusions regarding the general theoretical model
The general theoretical model
Researchers have worked hard in recent years to keep pace with developments in online technologies and
with children and young people’s online activities - activities which have earned them the title of ‘digital
natives’ by contrast with their ‘digital immigrant’ parents and teachers (Prensky, 2001). A few years ago,
there was little published research on children and the internet (Livingstone, 2003). Today, a dramatic
expansion in research is generating a growing consensus regarding key conceptual claims.
!
First, researchers concur that access is a prerequisite for, but underdetermines, use. A child-centred
account locates new technologies in the context of children’s everyday life to understand what they
complement, displace or remediate (Bolter & Grusin, 1999), how they fit meaningfully in established
social, spatial and temporal routines (Bakardjieva, 2005), and how far they afford new opportunities or
risks (Hutchby, 2001).
!
Second, research should avoid a technologically determinist, impact-centred approach, and instead
seek to understand how the internet is socially shaped, in terms of institution, design and political
economy, and also meaningfully appropriated in diverse contexts by its users (Berker, Hartmann, Punie,
& Ward, 2006). This is not necessarily to assert a social determinism, but rather to ask careful questions
about the dynamic and contingent relations between users and technologies, and between practices of
social shaping and technology use.
!
Third, research should sidestep the simple polarisation of ‘real’ and ‘virtual’ or ‘offline’ and ‘online’ so as
to pinpoint their intersections and mutual influences. Similarly, it should avoid the moral panics that
characterise media coverage and, to some degree, public understanding (Oswell, 1999).
!
Fourth, it is vital to recognise that it is part of childhood and adolescence to experiment, take risks, push
adult-imposed boundaries, and so forth. Thus it is important that online risks are addressed not by
restricting access (for this may produce evasion of adult regulation) but by enhancing critical literacy
skills, and by understanding what makes children resilient or able to cope with risk (Coleman & Hagell,
2007; Frydenberg, 2003)
!
Fifth, the solid tradition of research on parental strategies for regulating their children’s television use is
being extended to the internet, but here the task is more complex (Valkenburg, 2004). Given the notable
gap in parental and child accounts of both risk and regulation (Staksrud, 2005), and given the growing
challenge posed by not only fixed but mobile and convergent online technologies, it is crucial to seek
the means of improving the effectiveness of parental regulation strategies.
17
Project website and access to EVS data from http://www.europeanvalues.nl/.
61
Within this broad framework, this report has examined in detail the body of European evidence on children
and young people’s online opportunities and risks. To do so, and to respect the above theoretical principles,
a general model was proposed at the outset in order to clarify the hypothesised relationships among key
variables (see chapter 1).
As shown in Figure 2.8, at the heart of the model are the intersecting variables: access, usage, attitudes and
skills and, central to our focus, risks and opportunities. This nexus of factors is social in character, shaped by
a range of contextual factors (such as use of the Internet at home or school) and mediated by the actions
and beliefs of parents, teachers and peers.
Figure 2.8: The general model of the research field
Online activities of children
Access
Age
Risks and
opportunities
Gender
Usage
Attitudes
and skills
SES/inequality
Mediation by parents, teachers and peers
Individual level of analysis
Each of these variables may, in turn, be influenced by or dependent on the child’s age, gender and
socioeconomic background. Based on both academic literature and policy assumptions, a series of research
questions and hypotheses were specified at the outset of this report, in order to examine the findings in a
systematic way.
Identifying cross-national commonalities
Focusing here on the individual (rather than the country) level of analysis, this report has examined the
extent to which this general model holds across Europe, and the results are here summarised and
discussed.
The same analysis has also revealed ways in which the findings vary by country. These are summarised in
section 2.5.2, in order to propose country classifications according to key variables and these, in turn, are
explained at the country level of analysis in chapter 3.
Reiterating once more that although a considerable body of empirical findings has been identified across
Europe, this remains patchy, inconsistent and often not strictly comparable, a series of cross-national
commonalities have been cautiously identified. These are summarised and discussed below in terms of the
numbered research questions and hypotheses that structure this overall report.
We begin with the central portion of Figure 2.8 (shaded dark), examining the relationships between access
and use, attitudes and skills, opportunities and risks.
62
Overall conclusions regarding children’s online access and use
Research question R2.1.1: What/how much access to the internet and online technologies do children have?
! Most reliable figures on internet access in Europe concern the adult rather than child population. These
reveal, nonetheless, both a general rise in access across all countries and persistent differences in
access (cf. the digital divide) across and within countries (for more details see chapter 3).
! However, access under-determines use (Livingstone, 2002), for a child may have access to the internet
without using it, both at home and at school. Hence this report has focused on usage figures for children.
Research question R2.1.2: How much use of the internet and online technologies do children make?
!
In terms of frequency of use, the evidence is consistent with Rogers’ diffusion curve for technological
innovations (Rogers, 1995), namely that as access diffuses across countries, usage follows, with first
the early adopters taking up the internet, then the mass market and finally the laggards catching up.
Countries vary in the stage of this process reached, marking a continuing digital divide across Europe.
!
As research on the domestic appropriation of technologies shows (Silverstone et al., 1992; Haddon,
2004), underlying this process is a considerable effort by parents and children. They must make sense
of the technology and its various hardware and software components, they must rearrange their homes
and daily timetables to fit internet use into busy lives, and they must work out in symbolic terms the
benefits or risks for their lives and their social relationships within and beyond the home. In these ways,
they render the internet meaningful and therefore ‘useful’ in social and cultural terms.
!
Across Europe, the differences in amount of use remain more striking than the similarities (inviting a
classification of countries by children’s use of the internet - see 2.5.2). However, in all countries, certain
common features are also evident, reflecting this general process of diffusion and appropriation.
!
Most notably, the evidence across Europe shows that the more parents use the internet, the more
children do so also. Given the widespread assumption that children are the digital natives and parents
the digital immigrants, this is a counter-intuitive and important finding.
!
To be specific, across Europe, children (under 18 years old) are, according to the Eurobarometer survey
of 2005/6 (EC, 2006), marginally more likely (50%) than adults in general (47%) to use the internet. But
they are less likely to the use the internet than are parents/carers in particular (65%).
!
The findings qualify this general conclusion in two ways, permitting a more refined picture than has
previously been possible. First, across most countries, the child’s age matters. For younger children (up
to 11 years old), parents are the greater users and, presumably, more skilled therefore, than children –
challenging simple assumptions regarding ‘children’ as ‘digital natives’.
!
For teenagers, however, the picture reverses: teenagers across Europe are more likely (87% of 12-17
year olds) to use the internet than are parents of teenagers (65%). Teenagers are, indeed, the digital
natives, therefore. The role of parental responsibility for children’s internet safety should be approached
differently for children and for teenagers, for this reason.
!
Second, there are some significant cross-national differences here, with children overall more likely to
use the internet than their parents in certain countries (Poland and Portugal; see 2.5.2).
!
In these countries, therefore, policy expectations that parents can, in practice, take responsibility for
their children’s internet safety, should be especially carefully qualified.
Hypothesis H2.1.1: Children whose parents use the internet are more likely to use the internet themselves.
!
The above discussion applies on a generational level: child users in Europe vs. parent users in Europe.
!
Further analysis of the Eurobarometer findings showed that a child is more likely to use the internet if
their own parent(s) are users (58% of their children use the internet) compared with if their parents are
not users (34% of children use the internet). This holds across all countries, though the comparison
between parent users and parent non-users necessarily cannot be made in those countries where
nearly all parents are users (Estonia, Finland, Sweden). In some countries, the difference between
children, whose parents are online, and those, whose parents are not, is particularly high; for example in
Italy the difference between these groups is more than 30 per cent.
63
!
Across Europe, the findings also show that if parents use the internet at home, the influence on their
children is even stronger: 61% of children whose parents use the internet at home also use the internet;
only 9% of children whose parents do not use the internet at home do themselves use it.
!
Since parents are more likely to use the internet at home than adults in general, and that parents of
teenagers are more likely to use the internet at home than parents of younger children, it may be
concluded both that parents use the internet in order to encourage their children and that parents use
the internet because they have been encouraged to do so by their children.
!
Whichever the direction of causality, it appears that parental use has a positive association with
children’s use. Assuming the causal direction is, at least partly, from parent to child, to encourage
internet use among children it would be worth encouraging parents also to use the internet themselves.
Research question R.2.1.4: Where do children in Europe use the internet?
!
Across Europe, children are equally likely to use the internet at home (34%) and at school (33%), with
other places of much lesser importance (friend’s house – 16%, someone else’s house – 5%, library –
4%, internet café – 3%, or elsewhere – 2%).
!
There is a positive correlation between use at home and school – the more children use the internet at
home in a country, the more they are likely to use it also at school. The reverse is also the case,
meaning that children in some countries are doubly disadvantaged.
!
Further analysis of the evidence regarding home and school use provides the basis for a country
classification based on location of use – see 2.5.2.
!
Also cross-nationally, there is some evidence that in countries with low public or domestic access,
children are relatively more likely to go to internet cafés (e.g. Bulgaria, Poland). As these tend to be
unsupervised, from a child protection point of view, this may raise concerns.
Overall conclusions regarding children’s online opportunities and risks
Research question R2.2.1: What are the main opportunities experienced by children online?
!
Across Europe, a fair body of research evidence suggests that adults and children agree that children
use the internet as an educational resource, for entertainment, games and fun, for searching for global
information and for social networking, sharing experiences with distant others. Other opportunities, such
as user-generated content creation or concrete forms of civic participation, are less common.
!
There is little cross-nationally comparable evidence regarding the incidence and take-up of these
various opportunities. Thus it seems that, once they gain access (and skills), children in all countries
prioritise online communication, various forms of entertainment and play, and information provision;
meanwhile for parents, the benefits of educational resources come higher on their agenda.
!
If online opportunities are to be increased across Europe, much depends on the child’s role (their
motivation and resources) and the online provision available to them (and, thus, the providers’ motives
or social goals). Thus conceptually, EU Kids Online offers a framework by which to classify online
opportunities in Figure 2.5, according to providers’ motives (education and learning, participation and
civic engagement, creativity, identity and social connection) and the child’s role (as recipient, as
participant, as actor). The resulting 12 cells are unevenly studied at present, with some key gaps in
terms of creativity, civic opportunities, online sources of help and advice, and so forth.
!
It is further proposed that each child climbs a ‘ladder of online opportunities’, beginning with informationseeking, progressing through games and communication, taking on more interactive forms of
communication and culminating in creative and civic activities. Though many variants are possible, one
implication challenges the popular assumption that communication and games playing are ‘timewasting’ for, instead, they may provide a motivational step on the way to ‘approved’ activities. This
proposal merits further research.
64
Research question R2.2.2: What are the main risks experienced by children online?
!
In a parallel framework for online risks, EU Kids Online classifies these according to the three modes of
communication afforded by the internet: one-to-many (child as recipient of mass distributed content);
adult-to child (child as participant in an interactive situation predominantly driven by adults); and peer-topeer (child as actor in an interaction in which s/he may be initiator or perpetrator); a second dimension
acknowledges four main forms of risk to children’s development and well-being - commercial,
aggressive, sexual and value threats – in terms of provider’s motives. It is noted that while the specific
risks that fall into each cell may change over time, the categories are more enduring.
!
A portrait of online risks across countries can be derived from the available evidence, but it is important
to note, first, that not all risks discussed in public and policy have yet been researched; also, evidence
of their incidence, distribution and possible consequences on a reliable cross-national basis, is sparse;
last, risks are particularly difficult to define in culturally-consensual ways, and they are difficult to
research in methodologically-rigorous and ethically-responsible ways.
!
Overall, the findings suggest that online risk attracts public concern and policy attention with
justification. In most countries, significant minorities and, in some cases, a majority of children,
especially teenagers, are encountering a range of aggressive and sexual risks. These include content,
contact and conduct risks.
!
These findings do provide the basis for a tentative classification of countries according to likelihood of
encountering online risks (see 2.5.2).
!
Looking across European countries, there are grounds for proposing a rank order of risks in terms of
overall incidence, as follows.
1. Giving out personal information: the most common risk – estimates around half of online teens, with
2. Seeing pornography: second most common risk at around 4 in 10 across Europe, but there is
considerable cross-national variation (25% - 80%).
3. Seeing violent or hateful content: third most common risk at approx one third of teens and a fair
degree of consistency across countries.
4. Being bullied/harassed/stalked – generally around 1 in 5 or 6 teens online, though there this is higher
in some countries.
5. Receiving unwanted sexual comments - only around 1 in 10 teens in one group of countries
(Germany, Ireland, Portugal) but closer to 1 in 3 or 4 teens in Iceland, Norway, UK and Sweden,
rising 1 in 2 in Poland.
6. Meeting a online contact offline – the least common but arguably most dangerous risk, there is
considerable consistency in the figures across Europe at around 9% (1 in 11) online teens going to
such meetings, rising to 1 in 5 in Poland, Sweden and the Czech Republic.
!
Several risks are still to be researched in comparative perspective – self harm, race hate, commercial
exploitation, and so forth.
!
In several countries, some measure of distress or feeling uncomfortable or threatened was reported by
15%-20% of online teens, this suggesting, perhaps, the numbers for whom risk crosses over into a
degree of harm.
!
Some of the high reports of risk – in Estonia, Poland, the Czech Republic – require urgent awarenessraising. Similarly, the advent of new forms of online activity – e.g. social networking – points to the need
for urgent new advice to children and young people. As estimates for now-familiar risks continue to be
substantial, these too require continued attention to keep them in children’s minds.
Hypothesis H2.2.6: Since most children make the broadest and more flexible use of the Internet at home,
they will also encounter more risk from home than school use (i.e. a question of the relation between access,
use and risk).
!
This hypothesis is particularly important in terms of directing safety awareness initiatives, whether
targeted towards parents or teachers, or whether framed in terms of home or school use.
65
!
Findings from the pan-European Eurobarometer survey suggest that, according to their parents,
children encounter more online risk through home than school use (though this may be because parents
know little of their children’s use at school).
!
However, among those children who use the internet in an internet café or at a friend’s house, these are
also risky locations, according to parents (especially compared with school use).
Research question R2.2.5: What is the relation between online opportunities and risks?
!
UK evidence showed a high positive correlation between number of online opportunities and number of
online risks for 12-17 year old internet users. This led the EU Kids Online researchers to seek similar
findings in other countries.
!
However, little further European research was found that examined the association between online
opportunities and risks. We can only suggest that European policy makers face the dilemma that
increasing opportunities tends to increase risks, while decreasing risks tends to decrease opportunities.
!
Future research must identify ways of increasing online opportunities for children while decreasing risks.
Overall conclusions regarding children’s online attitudes and skills
This report has examined four hypotheses regarding children’s online skills and attitudes. However, as each
examines the relations with age, gender and SES, these are addressed below, where findings for the
following hypotheses are presented.
!
H2.3.1: As children get older they gain greater online skills, including self-protection skills
!
H2.3.2: Children who use the internet longer and for more activities develop more skills
!
H2.3.3: There are inequalities in skills and literacies as a consequence of inequalities in SES
!
H2.3.4: There are gender differences in the levels of skills (higher for boys)
Also relevant here is one of our main research questions.
Research question R2.2.4: Is there evidence showing the consequences of online risks or evidence showing
how children cope with online risks?
!
Once exposed to risk, how do children respond? In psychological research, this question is being
framed in terms of adolescents’ development of ‘resilience’. Thus far, however, little is known of
children’s abilities to cope with, or their resilience towards, online risk. Here more research is needed.
!
On a pan-European level, 31% of parents in the Eurobarometer survey say their child has encountered
harmful content on the internet, and 66% of parents say their child knows what to do in such situations.
!
But, comparing across countries, there is a negative correlation between these indicators: the higher the
percentage of parents in a country who claim their child has encountered harmful content, the lower the
estimated ability of children in that country to cope with potentially harmful encounters, and vice versa.
!
Some caution is needed in interpreting this correlation. It may that in low risk countries, children have
learned to cope; but it may be that in low risk countries, parents are unaware of their need to cope and
so overestimate children’s abilities. Similarly in high risk countries, children may really be less able to
cope or, possibly, these parents are more aware of children’s need to cope. Research is needed that
follows children from risk exposure through to their coping strategies and to any consequences.
!
Qualitative research points to a series of strategies that children are developing to cope with online
risks. How these are applied, and whether they are effective, remains unknown, but the manner of
reporting suggests that children feel in control and confident in using these strategies. Since this
research is mainly qualitative, there is also a need for quantitative research here, to establish the
relative uses of different strategies across the population and among specific subpopulations (especially
those deemed ‘at risk’).
66
!
However, as noted above, knowing good strategies may not affect children’s actual practice in risky
situations, however, so interviews in which children evince good sense should not be treated
complacently. This is especially the case since, although there is evidence that children are developing
their own strategies to respond to online risk, many studies continue to report that only a small minority
tells an adult and so can receive adult guidance.
In the following section, we address the palest grey section of Figure 2.8, to examine in turn the influence of
the child’s age, gender and socioeconomic background on the above findings for access and use, risks and
opportunities, attitudes and skills.
The role of age in influencing children’s online activities
Hypothesis 2.1.2: As children get older their access to and use of the Internet and online technologies
increases.
!
The hypothesis was framed in this manner as it is widely assumed, and shown in some published
literature (Livingstone & Helsper, 2007; Livingstone, 2002; Staksrud, 2005) that older children use the
internet more. Sometimes, the conclusion is drawn that, as they are also more mature, they are less at
risk (see below). However, the available findings permit a more refined account.
!
Specifically, findings from the national reports identify a mixed array of age-related factors that influence
children’s online access and use. These suggest that, for one reason or another, a simple linear
increase in internet use as children get older may not apply. Particularly, in several countries, there may
be a peak in use in the mid teens.
!
Younger children use the internet less, it seems, because their parents are more restrictive and
because there is less content provided for them. Older teens (17+) may use the internet less than
younger teens because they have more alternatives available to them.
!
According to pan-European Eurobarometer data (based on parental reports), children’s use increases
until 12-13 years and then plateaus.
!
Comparisons by countries grouped according to overall amount of internet use suggest that in high use
countries, children get online younger than in low use countries (see 2.5.2).
!
The safety awareness implications of the finding that younger children who are online are likely to live in
high use countries have yet to be pursued – and this may change as younger children begin to go
online in low use countries. Similarly, in low use countries, online teens may be less experienced users
than their counterparts in high use countries.
Hypothesis H2.2.1: As children get older they are exposed to a greater amount and range of online risks.
!
Eleven of the 21 countries had evidence relevant to this hypothesis, and in eight the evidence was
supportive, in three it was contradictory.
!
On balance, it is concluded that older teenagers do encounter more online risks than do younger teens.
!
In France, Germany and The Netherlands, there was evidence that younger teens are more risk taking
than over teens, however.
!
Further, the phenomenon of young children using the internet is too recent for a strong evidence base,
although this raises new and pressing questions which should now be researched.
Hypothesis H2.2.2: As younger children gain online access they are increasingly exposed to online risk.
!
Public concern is sometimes expressed regarding younger children, now going online and so
encountering risks that they are insufficiently mature to cope with. Although this concern is countered by
the greater parental mediation received by younger children, there is too little evidence to conclude on
this point.
67
Hypothesis H2.3.1: As children get older they gain greater online skills, including self-protection skills.
!
It is hypothesized that as children get older, they gain greater online skills (or Internet literacy) and so
gain more opportunities and also, presumable, gain the skills enabling self-protection from online risks.
!
Findings were available in eleven countries. Overall, the evidence supports the hypothesis that skills
increase with age.
!
Measuring children’s online skills, whether through qualitative or quantitative methods, is particularly
difficult and so, for the most part, it is unclear if increasing skills results in an increasing ability to cope
with or avoid online risk. The research community is actively addressing this methodological challenge
now, but no consensus or clear measures have yet been produced.
!
Where evidence is available, it does seem that increasing skill may increase self-protection. But one
should not be complacent about these growing skills, as there is also some evidence that children
‘know’ how to act safely online but in practice they take risks nonetheless. In this area as in others, the
link from safety knowledge to behaviour change is often uncertain.
!
Given these qualifications, it may still be concluded that as children grow older their level of skill
increases, and this is likely to include their abilities to protect themselves from online risks.
!
The development of skills is part of a more complex picture, however. Frequency of internet use,
amount of time spent online and the range of activities all increase with age, as does confidence, all of
which impact on young people’s skills levels, perceptions of skill and expertise and their adoption of
safety strategies.
Hypothesis H2.3.2: Children who use the internet longer and for more activities develop more skills.
!
Although it would seem obvious that children who use the internet for longer and for more activities
would develop more internet-related skills and literacies, only three countries provided data for this
hypothesis. In Austria, Norway and UK, findings showed a positive correlation between frequency of use
and online competence and safety. However, with such few data, generalisations are unsafe.
The role of gender in influencing children’s online activities
Hypothesis 2.1.3: There are no gender differences in children’s access to or amount of use of online
technologies.
!
The published literature on gender differences is very mixed. Some research suggests that early gender
differences for the home computer no longer exist. Other research suggests that gender differences are
becoming less a matter of strong inequalities and more a matter of subtle differences in preference or
style (Livingstone & Helsper, 2007; Livingstone, 2002; Staksrud, 2005; Bird & Jorgenson, 2003). Hence
no directional hypothesis was framed a priori.
!
Overall, contradicting the hypothesis, most findings from the national reports point to a fairly consistent
pattern of gender differences across Europe. On balance, we can go beyond specific country results to
suggest that boys use the internet for more time and in more places, than do girls.
!
Yet there are indications that these inequalities are becoming less and more and younger children go
online. And in a few countries (Ireland, Netherlands, Portugal and the UK), the Eurobarometer survey
finds internet use to be greater among girls than boys.
Hypothesis H2.2.3: There are gender differences in the range/types of uses/opportunities.
!
Over half the countries had evidence that there are gender differences in children’s online activities, and
only a little evidence contradicted the hypothesis.
!
It is concluded that, across Europe, while both boys and girls enjoy a range of online opportunities,
there is clear evidence of gender differences in online activities and preferences. Girls prefer activities
that involve communication, content creation and collaboration. Boys prefer competition, consumption
and action.
68
!
As yet, too little is known regarding the relatively new phenomena of social networking, online and multiuser gaming and other web 2.0 activities. There may also be an interaction between age and gender
(with gender differences increasing through the teens) but more research evidence is needed here.
Hypothesis H2.2.4: There are gender differences in the range/types of risks.
!
Fourteen countries provided research results supporting the hypothesis and there was little evidence to
contradict it.
!
Overall, it was concluded that boys are more likely to seek out offensive or violent content, to access
pornographic content or be sent links to pornographic websites, to meet somebody offline that they
have met online and to give out personal information.
!
Girls are more likely to be upset by offensive, violent and pornographic material, to chat online with
strangers, to receive unwanted sexual comments and to be asked for personal information but to be
wary of providing it to strangers.
!
Both boys and girls are at risk of online harassment and bullying.
Hypothesis H2.3.3: There are gender differences in the levels of skills (higher for boys).
!
In studies where children/young people self-report and self-evaluate their internet skills, boys tend to
rate themselves higher than girls.
!
There is little evidence based on tests or objective examination of children’s skill levels. There is some
evidence in some countries that boys’ greater online confidence (or self-efficacy) may lead them to take
more, not fewer, online risks. It may also enable them to evade adult regulation.
The role of socioeconomic status in influencing children’s online activities
Hypothesis H2.1.4: There are inequalities in access as a consequence of inequalities in SES (socioeconomic
status e.g. household income, parental education, social class)
!
In almost all countries, there is evidence to support this hypothesis, for higher SES households are
more likely to provide their children with access to the internet, while lower SES households are less
able to do this. Only Iceland and Sweden show little evidence of such inequalities.
!
Different studies operationalise different aspects of SES – typically parental income and/or parental
education – requiring more research if we are to understand how such inequalities can be reduced in
future. Such factors as minority status, social capital, school type and so forth may also play a role, yet
to be clarified. This is partly a problem of sample size (minorities measured in a sample survey may be
represented by few people) and partly one of measures (indicators of inequality vary across countries).
Hypothesis H2.1.5: There are inequalities in online use as a consequence of inequalities in SES.
!
In most countries, there is evidence to support the hypothesis that children from higher SES homes
make greater or more frequent use of the internet.
!
It is not always clear, in these studies, whether the comparison is for all children, or just for those with
internet access. However, in some countries, there are few differences in amount of use, especially if
comparing high and low SES children who do have internet access.
Hypothesis H2.2.5: There are inequalities in use/opportunities as a consequence of inequalities in SES.
!
The well-established debate over the digital divide justifies the hypothesis of inequalities in children’s
use of and opportunities gained through the internet as a consequence of differences in SES.
!
Though evidence was lacking in several countries, there is general agreement throughout many
European countries that SES and the type of use or opportunities are correlated.
69
!
Generally, it appears that children from higher SES backgrounds make more use of the internet (in
terms of frequency/amount of use) and use it more for education/information/civic purposes. Lower SES
children may use the internet more for downloading music, leisure and entertainment.
Research question R2.2.3: Are there SES differences in children’s exposure to risk?
!
Across Europe, there was little evidence to be found on this question in most countries – hence it is
treated as an open question rather than by framing a hypothesis.
!
Insofar as findings are available, the evidence in each country points to a correlation between SES and
exposure to risks, with the exception of Iceland. Most of these findings concern content and contact
risks. In the main, it seems that lower class children are more exposed to risk online.
!
Since lower SES children already experience disproportionate disadvantages and, so, may
disproportionately lack the resources to cope with online risk, further research to confirm this conclusion
is a priority.
!
In terms of safety awareness, these findings suggest the value of targeting interventions at lower class
children especially.
Hypothesis H2.3.4: There are inequalities in skills and literacies as a consequence of inequalities in SES.
!
Since few studies examine either SES or skills, this hypothesis cannot be examined reliably.
Finally we consider the intermediate grey section of Figure 2.8, addressing the question of mediation.
The mediating role of parents in influencing children’s online activities
Unfortunately, there is too little evidence across Europe to comment on the mediating role of either teachers
or peers although, for good theoretical reasons, these are included in the overall model (Figure 2.8). In other
words, it is very likely that teachers and, especially, peers influence children’s online activities, just as they
influence many other aspects of children’s lives (Irvine & Williams, 2002; Lawson & Comber, 2000).
Nonetheless, we have no basis for pursuing these here. What follows, therefore, examines evidence for
parental mediation (or domestic regulation). For clarity, we also consider below how parental mediation is in
turn influenced by the child’s age, gender and socioeconomic status.
Research question R2.4.3: What are the main strategies of parental mediation practised?
!
Evidence on this question is available from 17 countries, though there is great variation in the evidence
available from the different countries, indicating continuing theoretical and methodological issues for the
research agenda.
!
Across Europe, it appears that time restriction is the most common strategy for parental mediation of
children’s online activities. Possibly there are financial reasons for this, but also it is relatively easy for
parents to implement in practice.
!
Parental supervision – by sitting with children or checking on them ‘over their shoulder’ – is important in
many countries also.
!
Parents claim that they also discuss online activities with their children but children are less likely to
report that their parents do this – there is often a gap between child and parental reports of mediating
strategies (SAFT, 2006; UK Children Go Online)
!
Although substantial minorities of parents in several countries appear to use filtering software, its
effectiveness remains unclear from the available evidence.
!
There is sporadic evidence of some further parental strategies, but these are inconsistently studied
across Europe.
!
It is generally evident that parents wish for more guidance regarding the management of their children’s
internet use.
70
Research question R2.4.3: Are there gender differences in parental mediation?
!
This question too has been relatively little examined. Where evidence is available, it appears that girls
are subject to more parental mediation than are boys. Since the findings reported above suggest that
boys are no less exposed to online risk, this finding suggests that safety guidance could usefully be
targeted to parents of boys.
Research question R2.4.4: Are there SES differences in parental mediation?
!
In countries where this question has been asked, it appears that there is more parental mediation in
higher SES families. Little is known that compares mothers and fathers. It is also possible that this
finding reflects a social desirability bias on the part of parents.
Hypotheses H2.4.1: As children grow into teenagers they are subject to reduced parental mediation in their
use of the internet.
!
The general picture is very clear, holding across Europe: parents impose fewer and less strict rules
concerning media, including internet, usage as their children grow into teenagers.
!
One interpretation is that this is a fair reflection of teenagers’ growing maturity and rights to privacy and
independence. On the other hand, there is more evidence regarding online risks encountered by
teenagers than younger children – and it may be that parents have simply given up on the attempt to
regulate children by the time they become teenagers.
Hypothesis H2.4.2: More parental mediation results in reduced exposure of children to online risks.
!
This is a crucial hypothesis – widely assumed by policy makers but rarely examined empirically by
researchers.
!
Unfortunately, there is a very little evidence to support or contradict this hypothesis because 17 countries
out of 21 reported that there is no pertinent evidence available.
!
The remaining countries reported conflicting findings. Ireland found evidence to support the hypothesis
that more parental mediation results in reduced exposure of children to online risks, while Poland found
evidence to contradict. United Kingdom found evidence both to support and contradict. Two countries
commented on a general basis; Cyprus` comment supports the hypothesis while Spain’s comment
contradicts it.
!
Here more than anywhere, we can only conclude that more research is needed.
Research question R2.4.6: Is there evidence that particular parental strategies or styles of mediation
effectively reduce the risk that their children experience online?
!
Again, this is a crucial question where findings are sporadic and inconsistent.
!
Some of the research literature, including from EU Kids Online (Livingstone & Helsper, in press),
suggests that more restrictive strategies are more effective. But these restrict use generally and so
reduce online opportunities as well as risks.
!
Ideally, parental strategies of discussion, shared use and increasing media/critical literacy would prove
optimal, but there is little or no evidence that this is the case as yet.
2.5.2
Conclusions regarding the classification of countries
The findings of the research as presented in the previous chapters indicate commonalities and differences
between the European countries with regard to children’s online use. As a last step of this chapter these
commonalities and differences are discussed regarding their suitability as indicators for an overall
classification of the countries.
71
Internet use
The likelihood of children and teenagers accessing and using the internet provides a major way of classifying
countries, since internet use is related to most of the other indicators. In some countries, the diffusion of the
internet has almost reached the entire population, whereas in other countries online services are still “new”.
Since the indicators developed in this chapter reflect children’s and young people’s use of the internet, the
Special Eurobarometer “Safer Internet” (EC, 2006) is taken as the empirical basis for this classification.
Based on the percentage of children online, three groups were defined indicating “high” (> 65%), “medium”
(> 40%) and “low” (< 40%) internet use (see table 2.17, column 1).
Table 2.17: Indicators for country classifications regarding children’s internet use
1)
Children’s
internet use
Netherlands
Denmark
Estonia
Norway**
Iceland*
Sweden
Belgium
United Kingdom
Czech Republic
Slovenia
France
Austria
Germany
Poland
Ireland
Portugal
Spain
Italy
Cyprus
Bulgaria
Greece
High
High
High
High
High
High
High
High
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Low
Low
Low
Low
Low
Low
2)
Parents’
internet use
High
High
High
High
High
High
High
Medium
Medium
Medium
Medium
High
High
Low
Medium
Low
Low
Medium
Low
Low
Low
3)
Use at home
or at school
Home
Home
=
Home
Home
Home
Home
School
School
Home
=
=
Home
School
=
School
Home
Home
=
School
School
4)
Relevance of
use at home
High
High
Medium
High
High
High
High
Medium
Medium
High
Medium
Medium
High
Low
Medium
Low
Medium
Medium
Medium
Low
Low
less than 18 years. Definitions:
1) Basis: % children (0-17 years) who use the internet at any place; high: >64%, medium: >40% and < 58%, low: <39%.
2) Basis: % internet users whose parents also use the internet; high: >80%, medium: >65% and <80%, low: < 65%.
3) Basis: % children who use the internet at home and at school; ‘home’: use at home is more often (at least 3%) than at
school, ‘school’: use at school is more often (at least 3%) than at home,’ =’: use at home and at school are almost
equally distributed.
4) Basis: % child internet users who use the internet at home (and elsewhere); high: >80%, medium: >58% and <70%,
low: <50%.
The second classification (column 2) is based on whether the parents of children who are online also use the
internet themselves. Three groups are distinguished - countries where most children who use the internet
have parents who are online themselves, and countries in which almost one half of the parents of young
online users do not use the internet; plus countries in a mid-way position. The relevance of this classification
concerns parents’ competence to assist or support their children when they use the internet.
The third classification (column 3) is based on the observation that, in some countries, children use the
internet most often at home, whereas in others the school is more important for internet access. The fourth
classification (column 4) is based on the percentage of children who use the internet at home (and other
places) versus the percentage of children who are online at school (and other places) but not at home.
Several observations may be made at this point.
72
18
!
Country classifications based on columns 1 and 2 are almost but not entirely the same : there are some
differences, with Austria, Germany, and Italy having more online parents than might be expected (given
children’s use), and Poland and the UK showing the reverse feature.
!
High use among both children and particularly parents is generally associated with a relatively greater
reliance on home use, while lower use is also associated with a greater reliance on school use, in
general. However there are some exceptions, with the Czech Republic and the UK combining high or
medium use with a relatively greater reliance on school use, and Italy and Spain where low use is
combined with greater reliance on home use.
Online risks
A further group of classifications has been derived from risk related issues. Based on national studies (as
provided in the country reports), a classification of the countries according to the general likelihood was
st
proposed in chapter 2.2 (see table 2.18, 1 column).
This classification can partly be validated by the Eurobarometer survey. The parents’ perceptions, whether
nd
19
their child has encountered harmful content (table 2.18, 2 column) lead to a similar classification : the
majority of countries for which both data are available belong to the same group. Thus, the two sources
confirm that Poland and the Czech Republic are high risk countries, and that France, Germany and Italy are
low risk countries. There are also some deviations (e.g. for the UK there is contradictory evidence).
Table 2.18: Indicators for country classifications regarding risks
Austria
Belgium
Bulgaria
Cyprus
Czech Republic
Denmark
Estonia
France
Germany
Greece
Iceland*
Ireland
Italy
Netherlands
Norway**
Poland
Portugal
Slovenia
Spain
Sweden
United Kingdom
1)
General likelihood
of risk experiences
2)
Parents’ perceptions of
likelihood of harmful experiences
3)
Parents’ perception of
ability to cope with risks
Medium
Medium
Nd
Nd
High
Medium
Fairly high
Low
Low
Nd
Nd
Medium
Low
High
Fairly high
High
Nd
Nd
Nd
Medium
Fairly high
Medium
Low
High
Low
High
Medium
High
Low
Low
Medium
Nd
Low
Low
Medium
Nd
High
Medium
High
Medium
High
Low
High
High
Low
High
Medium
High
Low
High
High
Low
Nd
Medium
High
High
Nd
Medium
Low
Medium
Low
Medium
High
1) Basis: country reports on national studies concerning children’s risk experiences (see above, chapter. 2.2).
2) Basis: Eurobarometer 64.4, parents’ answers for whether they think their child has encountered harmful content when
using the internet (see table 2.12; basis: parents of children who use the internet). “High”: >45%, “Medium”: <45% and
>30%; “Low”: <30%.
3) Basis: Eurobarometer 64.4, parents’ answers for whether they think their child is able to cope with situations, which
make them feel uncomfortable (see table 2.2.8). “High”: >66%, “Medium”: <66% and >51%; “Low”: <51%.
18
The Pearson correlation between the two indicators across 19 countries is r = 0.90.
The Spearman correlation between the ranking derived from the national reports (column 1, high=1, fairly high=2,
medium=3, low=4) and the percentage of parents who say their child has encountered harmful content (basis for column
2) is rho = -0.63.
19
73
The third indicator focuses on children’s ability to cope with risks – as perceived by their parents (see table
rd
20
2.18, 3 column). There is a negative correlation between risk and the ability to cope with it across
countries indicating that the higher the percentage of parents who claim their children have encountered
harmful content, the lower the estimated ability of children to cope with these potentially harmful encounters.
According to these indicators, Estonia and Bulgaria are the highest risk/lowest coping countries followed by
Poland and the Czech Republic – clearly a priority focus for future safety awareness initiatives. On the other
side of the spectrum, a group of seven countries (Belgium, Cyprus, France, Germany, Ireland, Italy, and the
UK) combine low risk and a high ability to cope.
Some caution is needed in interpreting these findings. It may be that in low risk countries, children have
indeed learned to cope; but it may also be that in low risk countries, parents are unaware of the need to cope
and so overestimate their children’s abilities. Similarly in high risk countries, children may really be less able
to cope or, possibly, in high risk countries parents are more aware of their children’s need to cope.
Parental mediation
The third part of classifications refers to different patterns of parental mediation. Table 2.19 proposes three
classifications based on the findings with regard to the parents’ rules about their children’s media use as
provided by the Eurobarometer “Safer Internet” survey.
st
The first classification (see 1 column of table 2.19) has been defined on the basis of the percentage of
parents who claim they have set rules about their child’s use of any media. This classification reflects
differences between the countries regarding parents’ tendency to regulate their children’s media behaviour.
The second classification indicates the percentage of parents who have set particular rules for the internet
and thus reflects the extent to which the internet is regarded as a medium, which needs regulation. To
rd
provide a simple indicator regarding the relative importance of TV and internet regulation, the 3 column
shows, for which of the two media parents are more likely to set rules.
Table 2.19: Indicators for country classifications regarding parental mediation
1)
Rules set for any
medium
(%)
Netherlands
Sweden
Ireland
France
Spain
Austria
Belgium
Germany
United Kingdom
Estonia
Czech Republic
Italy
Poland
Denmark
Cyprus
Greece
Portugal
Slovenia
Bulgaria
Iceland*
Norway**
20
High
High
High
High
High
High
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Low
Low
Low
Low
Low
Nd
Nd
Nd
2)
Rules set for the
Internet
(%)
High
High
High
Medium
Medium
Low
Medium
Medium
Medium
Low
Low
Low
Low
Low
Low
Low
Low
Low
Nd
Nd
Nd
3)
Comparison between
rules set for TV and the
internet
Internet
Internet
=
TV
TV
TV
=
=
=
Internet
=
TV
TV
Internet
=
=
TV
TV
Nd
Nd
Nd
The Spearman correlation across 19 countries is r = -0.52.
74
1) Basis: Eurobarometer 64.4, parents’ answers on whether they have set rules about their child’s use of any medium
(TV, mobile phone, games console, internet, computer; see table 2.15; basis: parents of children who use the internet).
“High”: >70%, “Medium”: <70% and >60%; “Low”: <60%.
2) Basis: Eurobarometer 64.4, parents’ answers on whether they have set rules about their child’s internet use; see table
2.15; basis: parents of children who use the internet. “High”: >50%, “Medium”: <50% and >40%; “Low”: <40%.
3) Basis: Eurobarometer 64.4, comparison of parents’ answers on whether they have set rules for TV and for the internet
(in per cent) think their child is able to cope with situations, which make them feel uncomfortable (see table 2.15). “TV”:
percentage for TV at least 5% higher than for the internet; “Internet”: percentage of Internet at least 5% higher than for
TV; = difference between TV and the internet smaller than 5%.
These classifications lead to the following observations:
!
Although there is a slight trend that the likelihood of parental rules is higher in Northern and Western
countries, Denmark with a low percentage of parents who have set rules, and Spain with a high
percentage for parental regulation are significant exceptions.
!
Being one of the parts of the first indicator, it is plausible that the second indicator leads to a very
21
similar classification. However it is interesting to note that some countries deviate from this pattern:
In France, Spain, and particularly in Austria the likelihood of internet regulation is lower than might be
inferred from the strong tendency to regulate the use of other media.
!
The third classification, which indicates whether parents think, the internet needs more regulation than
TV (or the reverse) is quite independent of the general tendency to set rules. Some of the ruleoriented countries put more emphasis on TV (Austria, France and Spain), some of them (the
Netherlands, Sweden) are more likely to regulate the internet. It is obvious that the relevance of
22
internet regulation is highest in high use countries and lowest in low use countries.
An overall classification of countries
Based on the above three kinds of classifications an overall classification shall be developed. Given the
correlations between the different indicators, which have been mentioned so far, it seems advisable to take
the general likelihood of children’s online use as the first dimension (see table 2.20). If one takes the
classification of risks as derived from the national reports as the second dimension almost all cells of the
cross table have at least one country; the only exception is that there is no country with high internet use and
low risks.
Table 2.20: Overall country classification on the basis of online use and risk perception
Children’s internet use
Online risk
Low
Medium
Medium
Cyprus
Italy
Greece
Portugal
Spain
France
Germany
Austria
Ireland
High
Bulgaria
Czech R.
Poland
Slovenia
Low
21
High
Belgium
Denmark
Sweden
Estonia
Netherlands
Norway
UK
The Pearson correlation between the two underlying indicators across 18 countries is r=.75.
The Pearson correlation between the difference of rules for TV and rules for Internet on the one hand and the
percentage of internet users is r=-.63 for children’s online use and r=-.74 for the parents’ online use. As said earlier it has
to emphasized that these figures are based on parents whose children use the internet, so there is no direct influence of
the simple fact that children are more likely to be online on the parents’ tendency to set rules. Rather it can be
concluded, that parents in high use countries are more aware of online risks and thus set more rules.
22
75
The above table suggests that:
!
High use of the internet is rarely if ever associated with low risk.
!
Low use of the internet may be associated with high risk.
!
High use, high risk countries are, for the most part, wealthy Northern European countries.
!
Medium use, high risk situations are characteristic of new entrants to the EC.
!
Southern European countries tend to be lower in risk, though there are differences among them.
The proposed classifications of countries reflect different conditions for children and teenagers with regard to
online risks and opportunities. In order to explain and to better understand these differences, the following
chapter will now turn to the country level of analysis and collect information on relevant contextual factors.
76
3. Explaining differences and commonalities
between countries
Chapter 2 of this report has described the results of the collection and comparative analysis of existing
research on children’s and teenagers’ use of the internet and online-related opportunities and risks in
Europe. Following the general model of the research field (see figure 1.1, chapter 1), the following part of the
report sets out to define and collect relevant contextual factors or background variables, which help to
explain the similarities and differences between countries outlined in chapter 2.5.2. As a result of theoretical
considerations we have identified six areas, which build the relevant contextual framework for children’s and
teenagers’ online behaviour:
!
Media Environment: This area includes the aspects of internet and broadband diffusion, internet safety
tools, and media content for children.
!
Internet regulation and promotion: this section will the extent to which the governments of the European
countries try to regulate the internet or ICTs in general; a particular focus will be put on the role of the
government and the regulator(s) on the one hand and the influence of NGOs on the other hand.
!
Public discourses: It is assumed that perceived risks and opportunities will partly depend on the public
discourse. This will be examined with respect to media coverage of children and internet, the role of
NGOs and related stakeholders in shaping public discourses, and specific key events, which might
frame the public discourse in some countries.
!
Values and attitudes: The hypothesis here is that cultural values and attitudes will shape parents’ as
well as the whole society’s perspective on the opportunities and risks of online media.
!
Educational system: This area includes aspects like the general literacy of the population, the education
of the parents’ generation, the kind of education for today’s children, the technical infrastructure of
schools, and existing approaches to internet related media education.
!
General background factors: Finally we will deal with some single aspects, which are relevant for
children’s and teenagers lifeworlds, such as the perceived levels of social change; the enthusiasm of
the government and/or the public about changes associated with the Information Society and the
general situation of free speech and censorship; societal structures as marked by inequalities,
urbanisation, work and social class, and migration and cultural homogeneity; the supposed role of the
state regarding questions of safer internet; the language situation; and finally the current status of what
has been analysed as bedroom culture.
Looking at the fields mentioned here it is obvious that the EU Kids Online network cannot do its own
research on each of these aspects. Although there have been strong efforts to encourage comparative
research on the European level on many fields, it is still extremely difficult to get a systematic comparative
overview on any of these aspects. Thus, given the fact that within the research design of EU Kids Online
these contextual factors shall serve as indicators for an exploratory analysis of contextual influences on
online opportunities and risks, we followed a highly pragmatic approach including two kinds of sources: a)
Whenever available we used internationally comparative statistics or classifications for the respective field;
b) in addition we built on national reports from the members of the EU Kids Online network, in which they
summarized the national evidence on the respective issue.
The following sub-chapters for each of the areas mentioned above are structured according to the following
questions: 1) What kind of information is available? 2) What are the key commonalities and differences to be
observed in Europe? 3) Is there evidence, which supports a classification of the European countries? 4)
Which hypotheses can be developed with regard to the explanation of differences and commonalities in
children’s and teenagers’ online experiences, which have been elaborated in chapter 2?
77
3.1 Media Environment
3.1.1
Internet and broadband diffusion
23
3.1.1.1 Sources of information
24
EUROSTAT provides comparative data on internet and broadband diffusion . The Digital Access index is
better in combining a variety of measures, but it is severely outdated, as it was last published in 2005/06
(ITU, 2006). For a more complex index the Network Readiness Index (NRI) from the Global Information
25
Technology report from the World Economic Forum (Network Readiness Index (NRI), 2008) is helpful.
However, in general the number of internet users is fairly well correlated with the results of the index.
3.1.1.2 Commonalities and differences between the countries
Internet diffusion varies across Europe ranging from 91% of population using the internet in Iceland to 34%
in Bulgaria (see table 3.1). In general, EU Kids Online members can be divided in three groups – those with
high internet diffusion, middle internet diffusion and low internet diffusion. High internet diffusion is most
common among Nordic countries and also Benelux countries and UK. The internet is used less among
Southern European countries and the former Eastern block – Greece and Bulgaria having the lowest
proportion of households connected to the internet and also the highest cost of internet access. According to
the cost of broadband use, there are two abnormalities in an otherwise fairly linear relation: Italy has very low
prices for internet broadband connections, but also low usage, whereas Estonia has one of the highest costs
of internet usage among EU Kids partners, but at the same time, usage is very high.
Table 3.1: Indicators for internet and broadband diffusion
1) % of
Internet
users
EUROSTAT
2007
Austria
Belgium
Bulgaria
Cyprus
Czech Rep.
Denmark
Estonia
France
Germany
Greece
Iceland
Ireland
Italy
Netherlands
Norway
Poland
Portugal
Slovenia
Spain
Sweden
UK
69
69
34
41
52
85
66
66
75
36
91
61
41
86
87
49
42
57
55
82
75
Ranking
among EU
Kids online
countries
according to
column 1)
9
8
21
19
15
4
11
10
7
20
1
12
18
3
2
16
17
13
14
5
6
Position
among
EUKIDS
project
according to
column 1)
2) % of
households
connected
to Internet
EUROSTAT
2007
Middle
Middle
Low
Low
Low
High
Middle
Middle
High
Middle
High
Middle
Low
High
High
Low
Low
Middle
Low
High
High
23
Author: Pille Pruulmann-Vengerfeldt.
Find EUROSTAT statistics at: http://epp.eurostat.ec.europa.eu/.
25
http://www.insead.edu/v1/gitr/wef/main/home.cfm.
24
78
60
60
19
39
35
78
53
49
71
25
84
57
43
83
78
41
40
58
45
79
67
3) % of
households
connected to
broadband
Internet
EUROSTAT
2007
46
56
15
20
28
70
48
43
50
7
76
31
25
74
67
30
30
44
39
67
57
4)
Position
in NRI
07/08
15
25
68
41
36
1
20
21
16
56
8
23
42
7
10
62
28
30
21
3
12
5) Lowest
cost of
broadband as
% of monthly
income
(2006)
0,11
0,04
1,07
0,35
0,16
0,08
0,52
0,01
0,02
0,65
0,03
0,09
0,01
0
0,04
0,21
0,07
0,1
0,23
0,01
0,02
3.1.1.3 Indicators for classifications of the European countries
The statistics shown above provide a meaningful and easily measurable and comparable basis for country
classifications. For the percentage of internet users (2007) the respective classification of the EU Kids Online
countries is shown in Table 3.1.
3.1.1.4 Hypotheses regarding the influence of this contextual factor on Safer Internet issues
It is highly plausible to assume that the general diffusion of the internet in the European countries explains
the substantial differences in the percentage of children being online, which have been described in chapter
2.1. Actually, the percentage of (adult) internet users as provided by EUROSTAT for 2007 is very highly
correlated (r=.89) with the percentage of children who are online, which has been based on the Special
Eurobarometer 250 from December 2005. The above classification of the countries involved in the EU Kids
Online network is highly similar to the classification based on children’s internet use; 13 out of 19 countries
fall in the same group, the remaining six countries fall into the neighbouring group.
Thus, the general diffusion of the internet is a strong factor influencing children’s use of the internet. With
respect to this factor, differences between the European countries are still massive. As a consequence, for
children in countries in which the internet diffusion has reached an advanced stage, online services are a
normal part of their media environment and everyday life, whereas for children in other countries it is
something that needs a specific effort and makes a difference to other children.
One would assume that the more common the internet is in a country, the more awareness there is of
internet safety issues. One would also assume that the cheaper the cost of the broadband connection, the
more common is internet use, especially among households with more deprived children such as those living
in a single parent household.
3.1.2
Internet safety tools
26
The most important question to be answered in this section is the extent to which Internet Service Providers
(ISP) offer internet safety tools (e.g. filters) or provide warnings/advice.
No reliable data are available for comparative purposes. Most of the data available here are based on
personal impressions after carrying out some non-exhaustive search about the state of the art regarding
internet safety tools in each of the countries. Moreover, almost none of the countries involved seem to
possess statistical data on internet safety tools.
On the basis of the national reports, the main findings can be summarised as follows:
In the majority of countries studied only the major ISPs provide (or at least advertise that they provide) safety
packages that include a wide range of services such as antivirus and anti-spyware protection, defence
against phishing attacks with URL filtering and anti-spam functions, detection of Wi-Fi intrusion, improved
personal firewall preventing intrusions by hackers and blocking networks viruses targeting loopholes in the
network, among other things. Most of these services work by performing regular scans on computers,
warning about security weaknesses in the operating system/browser, controlling the local network, and by
providing spam filtering. One disadvantage of these packages is that they are not free and therefore users
interested in these services must pay for them.
Of all the countries studied, only in Bulgaria was filtering software not popular, probably because these
services are not widely advertised in this country while ISPs do not provide many warnings or advice about
them. Therefore, there is an evident lack of awareness in the mainstream population about using these
filters. Finally, the acquisition of these packages or even of (legal) antivirus software is hindered by their high
costs.
Apart from offering the typical “safety “ packages, many ISPs also offer advanced parental control functions
that enable users to manage children's use of the web, blocking unsuitable sites and content. Many of these
packages also include the possibility of creating different settings for different family members. Some ISPs
even advertise that parents can disable the folllowing: gambling, hate speech sites, sites about drugs, adult
26
Authors: Verónica Donoso (Belgium), Anna Van Cauwenberge (Belgium)
79
sex sites, weapons or web-mail. In some countries such as in the case of France, the installation of the
parental control tool is the default option.
Only some ISPs throughout Europe provide additional information about children and the internet and on
how to protect children against online risks – although never on their home page (it requires some searching
to find this information). Indeed, most of the information provided by the ISPs relates to their own products,
whereas a very limited number provide some detailed information about children's safety on the internet and
parental regulation – information that goes beyond the company's products.
In some countries such as in Belgium, Ireland, Iceland and Slovenia internet service providers (mainly
through ISP national associations) participate in projects to safeguard online safety for minors. In Belgium,
for example, the Internet Service Providers Association (ISPA) is involved in several projects such as:
StopChildPorno (an informative site that is the national and civil hotline to report child abusive images found
on the internet), Delete Cyberhate (a website that serves as a national hotline for reporting illegal hate
speech on the internet), Safer Internet Belgium (a project targeted to raise awareness about the risks of
internet for children, parents and teachers) and Spamsquad (a site where internet safety tools are offered
and where also warnings/advice are provided).
In other countries such as in Ireland, the Internet Service Providers Association (ISPAI) operates the
www.hotline.ie service. This hotline combats illegal child pornography on the Internet and provides a secure
and confidential environment where children as well as adults may anonymously report pornographic content
encountered on the Internet. Whilst the primary focus of the Hotline remains Child Pornography, other forms
of illegal material do exist on the Internet and may be reported using this service. Also in Iceland and in
Slovenia most ISPs are very active in internet safety issues, besides offering filters and safety tools they also
cooperate extensively with national internet awareness node Safe-si (Insafe). They also collaborate in
projects to raise public awareness of online risks. Most Icelandic ISPs cooperate with the Icelandic node of
the EC Safer Internet Action Plan, SAFT (www.saft.is) helping to produce awareness material and funding
media campaigns among other initiatives. In Denmark the ISPs cooperate through the Danish IT Industry
Association also regarding promoting safety for children online and on various digital platforms such as
mobile phones. They consider all kinds of risks such as pornography and commercial exploitation. The ISP
members in the EU Kids Online advisory panel claim to be more restrictive in their own administration of
children’s access to services and content than the legally decided restrictions.
In some European countries industry also plays a key role in raising awareness. As a matter of fact, in
countries such as Greece, Italy, Ireland, Belgium, Estonia, Slovenia and the UK, companies such as
Microsoft, Norton Utilities, Yahoo, Vodaphone and Bebo offer some guidance and/or products regarding how
to improve safety and security online. In particular, in Italy, Belgium, Estonia and UK Microsoft local websites
offer safety advices for parents and/or educators signalling its own filters and teaching them how to apply
them. In the UK, Microsoft provides safety awareness training materials to every secondary school in the
country while in Estonia, Microsoft, in cooperation with local organisations (the Family Centre), has promoted
projects and research to identify risks related to children, to help distribute information about safety on the
Internet, etc. Other examples of industry support to online safety are found in Ireland, where the largest ISP,
Eircom, has recently introduced an e-Security package in conjunction with Norton Utilities, providing antivirus and firewall protection, identity theft support, email anti-virus and spam blocker, anti-spyware protection
and anti-phishing protection. Finally, Yahoo’s new online safety information and Bebo’s recently launched
safety site in the UK also illustrate how the industry can support online safety awareness among the
population.
Finally, it seems that only in a few countries, the government (through entities such as the Ministry of
Communication in Italy, the Belgian Privacy Commission or the Commission for the Protection of Minors in
Electronic Media and the State Media Authority in Germany) plays an important role in protecting minors and
in raising awareness of online risks by means of supporting initiatives, passing laws and developing projects
that promote minors’ safety on line.
Based on the country reports, one can meaningfully distinguish between three groups of countries when it
comes to the influence of ISPs on safeguarding safety online for children. However, although this
classification seems useful, the information provided in the national reports may not be enough to provide an
accurate classification of the countries involved in our project.
80
Table 3.2: Country variation in ISPs’ activity in safeguarding online safety
ISPs play an active role in
safeguarding safety online
ISPs offer the typical
paid “safety packages”
Belgium
Greece
Estonia
Germany
Ireland
Italy
Slovenia
the UK
Denmark
The Netherlands
Austria
Cyprus
Czech Republic
France
Iceland
Norway
Spain
Sweden
Portugal
ISPs provide (almost) no warning,
advice or information on safety issues
for children
Bulgaria
Incomplete information to perform classification for: Poland
The three categories can be described as follows:
1.
ISPs play an active role in safeguarding safety online for children by offering the typical “safety
packages” but also by participating in local projects to raise public awareness, by collaborating with
safety nodes, by producing and distributing online safety awareness-raising material for schools, etc.
2.
ISPs offer the typical paid internet “safety packages”, i.e. but are not actively involved in safeguarding
safety online for children. “Safety packages” typically include a wide range of services such as antivirus
and anti-spyware protection, defence against phishing attacks with URL filtering and anti-spam
functions, detection of Wi-Fi intrusion, improved personal firewall preventing intrusions by pirates and
blocking networks viruses targeting loopholes in the network, among others. Most of these services
work by means of performing regular scans on computers, warning about security weaknesses in the
operating system/browser, controlling the local network, and by providing spam filtering. One
disadvantage of these packages is that they are not free and therefore, users interested in these
services must pay for them.
3.
ISPs (almost) do not provide warning, advice or information on safety issues for children.
The presence of information and guidelines on online safety in ISPs’ websites raises public awareness
regarding online safety issues.
The presence of information and guidelines on online safety in ISPs’ websites has a positive effect on
children’s behaviours and attitudes regarding on-line safety issues.
The presence of information and guidelines on online safety in ISPs’ websites reduces children’s exposure to
risk.
Due to the lack of comparable data on safety awareness in the different countries these hypotheses cannot
be examined directly; comparing the above classification with the differences between the countries in terms
of risk perception does not result in a clear pattern. Although it is highly plausible that safety information
provided by ISPs can raise awareness and reduce risks, there is no concrete empirical evaluation available
so far.
81
3.1.3
Media content for children
27
There are no international statistics on this topic and the country reports within this study provided rather
diverse kinds of information. Only Germany, Italy and The UK provided some quantitative data. This points to
an urgent need to develop efficient procedures that would help to assess the range and quality of median
content for children.
In most of the countries studied, the Public Service Broadcaster seems to be the major media content
provider for children. However, when narrowing the scope to online content for children, some commercial
broadcasters turn out to also be important content providers. The latter usually offer a mix that includes a
limited range of national formats and a broad range of international formats. Within these international
formats, some popular TV channels that also provide online content include Nickelodeon and Disney.
One has to be careful when making general claims about the quality and variety of on-line media content
aimed at children. However, in countries such as Austria, Denmark, Germany and the UK, the media content
for children does seem to be rich and broad. To illustrate this, Austria’s public service broadcaster offers
several online sections for all children’s programmes with different thematic priorities: news, action, stars,
technology and science, animals and nature, quizzes, shows and television, etc. In Denmark the National
Danish Broadcasting company in particular but also the national TV2 channel have a strong focus on
children and adolescents and provide a good deal of services and content for all platforms: TV, radio, online,
mobile and crossovers. Commercial channels and institutions also provide substantial content. In Germany,
several programmes as well as websites for children (kika.de, kindernetz.de, tivi.de) are provided by the
public service broadcasters, as is the case in Sweden. As for the UK, the BBC is a strong offline (CBeebies,
CBBC) as well as online (BBC Children, BBC Learning, BBC Teens) content provider for children.
Because there was not enough information about the provision of online content for children in the national
reports, we decided to carry out a small “survey” in situ among the participants of the WP3 workshop in
Salzburg so as to determine the degree of significant online content provision for children. The results of our
survey are shown in the following table (table 3.3 and 3.4):
Table 3.3: Country variation in provision of positive online content for children
Significant positive online
content provision for
children
High
The Netherlands, the UK, Denmark
Between high and medium
Austria (because children there can also access German
websites); Ireland (because children can have access to
English language websites); Belgium (because they can use
Dutch websites)
Medium
Italy, Estonia, Norway, France, Belgium, France, Norway,
Germany, Iceland, Sweden, plus DK
Low
Spain, Greece, Cyprus, Portugal, Portugal, Poland, Iceland
27
Authors: Verónica Donoso (Belgium), Anna Van Cauwenberge (Belgium), Verónica Donoso (Belgium).
82
Table 3.4: Country variation in who provides online content for children
Major Online Content
Provider for Children
Public Service Broadcaster
Commercial media
Public and Commercial
Underdeveloped
Not mentioned
Austria, Ireland, Italy
Bulgaria, Cyprus, Estonia
Belgium, Denmark, Germany, The Netherlands, Norway,
Portugal, Slovenia, Spain, Sweden, The UK
Greece, Iceland
France
A classification according to the type and amount of significant positive online provision per country seems
meaningful. Even though this information is not explicitly available from the country reports, it would still be
important to explore not only the types and amount of online content available for children online, but also its
quality. From the point of view of 9-19 year old children, quality internet provision comprises, among other
things, good quality content addressing their interests and sites should be truly interactive including to
provide responses to their inquiries or contributions. Moreover, quality online content should also offer more
guidance not only in terms of content creation, but also in terms of safety issues such as improved protection
from unwanted content and attention to children’s privacy needs (UK Children Go Online). Following this, we
propose the following classification:
1. High significant positive online content provision for children: The amount and the types of online
websites that children can access is not only adequate in terms of its content and safety aspects, but it is
also abundant.
2. Medium significant positive online content provision for children. This means that even though there are
several high quality websites aimed at children, there are still many websites which do not provide
appropriate or sufficiently safe content.
3. Low significant positive online content provision for children: most of the available on-line content for
children is of poor quality and not safe enough.
Both classifications mentioned in the above tables (3.3 and 3.4) are meaningful regarding the EU Kids
Online project. The presence of a strong Public Service Broadcaster that is a (major) content provider for
children, offline as well as online, can play an important role in guiding and teaching children how to use the
Internet in a safe and constructive way. In contrast to commercial media, Public Service Broadcasters have a
particular responsibility to fulfil regarding the provision of quality content for children. As the internet
becomes more and more central in the world of children, Public Service Broadcasters have a crucial role to
play.
The provision of positive good online content for children has a positive effect on children’s behaviour and
attitudes regarding on-line safety issues.
The provision of good online content for children reduces children’s exposure to risk.
Although these hypotheses are highly plausible, it must surprise that there is almost no empirical evidence –
even on the national level – evaluating the effects of dedicated online content, which sets out to support
children in using the opportunities and avoiding the risks of the internet. In this respect there is a particular
need for additional research.
83
3.2 Internet regulation and promotion
3.2.1 Legislation and Policing (Regulation)
28
Comparative information on internet regulation can be drawn from the World Economic Forum (see chapter
3.2.1.3), whereas the information provided in the national reports are difficult to compare.
The European framework, set on the level of the European Union, is an important starting point. The EU has
built a regulatory framework for electronic communications based on five Directives.
!
The " Framework " Directive (EC, 2002a), which aims to promote competition, consolidate the internet
market for electronics communications and serve the interests of consumers and users.
!
Directive on the authorisation of electronic communications networks and services (the "Authorisation "
Directive) (EC, 2002b).
!
Directive on access to electronic communications networks and services (the "Access " Directive) (EC,
2002d).
!
Directive on universal service (the "Universal Service " Directive) (EC, 2002c).
!
Directive concerning the processing of personal data (the " Directive on Privacy and Electronic
Communications ) (EC, 2002e).
These directives have been incorporated or are going to be incorporated in the regulatory legislation of each
country. So we could assume that every country has or will have a basic common regulatory framework for
electronic communications, without prejudicing measures taken at national level. Within this these can
pursue general interest objectives, in particular those relating to content regulation and audiovisual and
telecommunication policy. We could say that this framework is oriented towards guaranteeing a more
competitive free market, but pays little attention to consumer and user interests, and no attention to the
specific protection of minors, except through the European Parliament “Recommendation on protection of
th
minor’s human dignity in audiovisual and information services” adopted on 20 December 2006.
On the other hand, the European Union passed a new directive about Audio Media Services in 2007 (EC,
2007), which replaces the “Television Without Frontiers” directive (EC, 1989), amended in 1997 (EC, 1997).
The latter directives (1989 and 1997) have been incorporated into the regulatory legislation of each country,
and the 2007 Act soon be implemented. This legislation assures the adoption of rules to protect the physical,
mental and moral development of minors as well as human dignity in all audiovisual media, including
audiovisual commercial communications. It has been a very important step because this Directive of 2007
covers all audiovisual media services (analogue and digital television, live streaming, web casting and NVO,
and video-on-demand), independently of the technology or distribution platform used (broadcasting television
– linear services – and on demand services – non-linear –). But this directive has still not been incorporated
into the national legislations.
Following the national reports it seems that there are for the most part few specific “computer crimes”, that is
crimes which are judicially considered to have as their object or instrument computer data and systems.
“Everything that is illegal in the real world is also illegal on the internet (such as child pornography, Nazi
content, fraud, etc) and will therefore be prosecuted”, is written in the Austrian Report, or “In Ireland, as well
as in most jurisdictions, what is illegal off-line is considered illegal on-line” is in the Irish Report, or “in
general, if something is forbidden in society it is forbidden on the internet, but overall there are few special
laws relating to the Internet” is in the Swedish Report.
However, there is a wide range of crimes in the National Criminal Codes which could be regarded as
“computer crimes”, as they are committed through the Internet: phishing, credit card frauds, bank robbery,
illegal downloading, industrial espionage, child pornography, harassment, bullying, scans, cyber-terrorism,
creation and/or distribution of viruses, spam and so on. All such crimes are related to and facilitated by
computers. Nevertheless, some countries have specifically referred to “computer crimes” in their legislation
over the last years. For instance, in 2007, France introduced the article 227-22-1 in its Criminal Code in
28
Author: Carmelo Garitaonandia, Maialen Garmendia
84
order to be able to prosecute any “sexual proposal to a minor under 15” using any electronic communication
means, and also the article 222-33-3 of the penal code condemns the recording or diffusion of images
related to “happy slapping”. In Ireland, a new offence of meeting a child following sexual grooming, on the
internet or otherwise, was included in the recently passed Criminal Law (Sexual Offences) (Amendment) Act
2007. In the UK, the 2003 Sexual Offences Act made grooming a child online for sexual purposes (in chat
rooms, email, instant messenger etc) illegal, and the grooming laws gained a high profile in the press; other
countries in the EC are now considering passing similar regulations.
In 2001, in Budapest, the European Council passed the “Agreement of Cyber-Delinquency” signed by all the
participating countries. This agreement classified computer crimes in four different categories, which can
also be seen in almost every National Criminal Code:
!
Crimes against confidentiality, integrity and the availability of data and computer systems,
!
Crimes of falsification and computer fraud,
!
Crimes related to contents, such as child pornography,
!
Crimes related to copyright.
The number of crimes committed on the internet has increased considerably over the last years and every
national police has a brigade, group or department which deals with online crimes. Fourteen European
29
countries are part of the European Working Party on Information Technology Crime , and they work very
closely with international organisms like Interpol and Europol. According to INTERPOL data, 50% of crimes
committed on the Internet are related to distribution, diffusion and the selling of child pornography, and some
European police forces work with the Child Exploitation Tracking System (CETS), a software system
developed by Microsoft which permits investigators to easily organise, analyse, share and investigate
information from its point of detection right through the investigation phase, and the arrest and management
of the offenders. This software is offered free by Microsoft, and the countries which have installed it are Italy,
the UK and Spain.
Probably, in every European country the laws which have the biggest influence on the regulation of contents
on the internet are contained in the Criminal Code, and some countries have reformed their Criminal Codes
and have incorporated the possession of child pornography (pornographic material such as photos, videos,
digitalised images, electronic files, etc., in which someone with the age of a minor has been involved) as a
30
crime and included other crimes related to children . The lower age limit at which pornography is considered
“child pornography” and, therefore, a crime, varies from country to country within the European Union. Whilst
Holland, Spain and Italy have the age limit as eighteen, some other countries such as Germany and Austria
set the limit at fourteen years of age. It is necessary to consider the fact that, very often, when a child
pornography crime is committed, some other crimes such as sexual abuse or rape can also be involved.
Table 3.5: Country variation in the age limit at which pornography is considered “child pornography”
Age
Countries
14
Germany,
Austria
15
Denmark,
Finland, France
16
Belgium, United
Kingdom
18
Greece, Holland,
Iceland, Italy,
Luxembourg, Portugal,
Spain, Sweden
Source: Report of the NGO Anesvad about child pornography on the Internet (Anesvad, 2003)
29
Austria, Belgium, Denmark, Finland, France, Germany, Italy, Netherlands, Norway, Portugal, Sweden, Switzerland,
Spain, and the United Kingdom.
30
For instance: the reform of the Spanish Criminal Code in 2003 (art. 187 and art. 189. The Organic Act, 25th November
2003), the amendments to the Bulgarian Criminal Code in 2002 were introduced incriminating child pornography as a
criminal offence with heavier sanctions than ordinary cases of pornography. Sanctions envisage imprisonment of up to 8
years, fines of up to 5,000 EUR and (in some cases) confiscation of property, and the reform of the new Portuguese
Criminal Code and Criminal Code Process in 2007. Regarding child protection new crimes were created as minor
pornography, the use of minor prostitution and genital mutilation is now expressly covered. Also, in case of crimes
practiced against persons under 16 years old, if the legal guardian of this person does not want to present charges, the
person can present charges themselves from the moment they are 16 years old until they are 18 and a half. Regarding
sexual crimes against minors, it is established that the crime does not prescribe before the minor is 23 years of age.
85
People who wish to wet up websites which include child pornographic content, try to get on servers in
countries where the legislation dealing with this matter is not as clear as in European countries. In fact,
servers in Brazil, some Latin American countries and former members of the USSR are often chosen.
Although half of the crimes committed on the internet are related to child pornography, there are many more
crimes in the National Criminal Codes for which computer systems are either the means or the object of
crime. For instance: exhibitionism and sexual provocation, prostitution and corruption of minors, threats, libel,
fraud, crimes related to the intellectual and industrial copyright, crimes related to the market and consumers,
etc. However, it is necessary to highlight that there are certain, very frequent behaviours on the internet that
are very difficult to prosecute because they has not been formalised in almost any of the national Criminal
Codes. Two very important examples of this are SPAM and port scan (a way of using other people’s IP
address and PC).
31
The World Economic Forum (2007) uses the Executive Opinion Survey (World Economic Forum, 2007) to
evaluate the laws relating to the use of information and communication technologies (e.g. electronic
commerce digital signatures consumer protection). The scale used in the survey is 1 = nonexistent to 7 =
well developed and enforced. Based on these scores, we can classify EU Kids Online countries in three
groups – Laws well developed and enforced, Laws averagely developed and enforced and Laws not that
well developed. Table 3.6 gives an overview of these classifications and scores are in the brackets behind
the country name.
Table 3.6: Country variation in the laws related to information and communication technologies
Laws well developed and
enforced
Denmark (6.01)
Estonia (5.90)
Germany (5.76)
Sweden (5.74)
Austria (5.70)
Norway (5.63)
UK (5.54)
Iceland (5.40)
Netherlands (5.39)
France (5.34)
Laws averagely developed and
enforced
Portugal (4.94)
Ireland (4.91)
Belgium (4.85)
Slovenia (4.82)
Spain (4.77)
Bulgaria (4.27)
Italy (4.27)
Czech Republic (4.21)
Laws not that well
developed
Cyprus (3.88)
Poland (3.69)
Greece (3.63)
Source: World Economic Forum 2007
Beyond that, based on the qualitative national reports, we can indicate that there are some countries where
in addition to the criminal code specific institutions regulate the internet. These countries and their regulatory
mechanisms are summarized in the following table (table 3.7).
Table 3.7: Country variation in the institutional regulation of the internet
Country
UK
Austria
Germany
Ireland
Italy
Others
Institution of statecontrol
X
X
X
X
Institution of selfregulation
X
X
X
X
Basic Criminal
Code
X
X
X
X
X
X
In most European countries which have inherited a tradition of freedom, there is no special control of internet
content. The criminal code is the law which determines if contents or a particular service is against the law.
Nevertheless, there are countries like Germany, the UK and Ireland in which the development of the internet
has been even faster than the average, and these countries have perceived and taken more seriously the
31
http://www.insead.edu/v1/gitr/wef/main/analysis/showdatatable.cfm?vno=2.18.
86
risks related to the internet and have created their own institutions of control. The case of Italy is probably
different, because while the development of the Internet in this country has not exceeded the average, Italy
has a tradition of controlling TV contents in order to protect children, so it has applied the same system of
control to the internet content.
On the other hand, Anglo-Saxon, Northern and Central European countries have a greater tradition of self
regulation than Latin and Southern European countries, in which legislation plays a more important role than
self-regulation. The former countries have created self-regulation institutions to help to control contents
aimed at children on the internet. Due to this, over the last years, countries such as Belgium, France,
Portugal, Spain and Bulgaria had to modify their Criminal Codes to include crimes related to the internet.
When we speak about institutions of State control, we are basically referring to institutions controlling online
contents for children, as probably every European country already has an institution with which to guarantee
market freedom on the telecommunications market (Office of Communications, Comisión del Mercado de las
Telecomunicaciones, Autorité de Regulation des Communications Electroniques et des Postes, etc) and also
an agency to protect data.
The classification from the World Economic Forum indicates that while about half of the countries perceive
themselves to have relatively adequate regulation on internet issues in general, there are still exceptions like
Cyprus, Poland and Greece where more regulatory mechanisms are needed. This seems to correlate with
other classifications fairly well, indicating that where there are more internet users, there is also more
legislation regulating activities on the internet. However, it must be kept in mind that this is not an absolute
scale, but rather perceptions of the adequacy of regulation. For example, although internet content is not
very regulated in Estonia, any kind of censorship is generally not well received by the public at large. Thus
laws can be regarded as being adequate in terms of public perception even if no actual content regulation
and child protection mechanisms are in place.
3.2.2
The role of government and regulator
32
The Networked Readiness Index (NRI, 2008) contains the component “Government success in ICT
promotion”. Government programmes promoting the use of ICT are compared on a scale (1 = not very
successful 7 = highly successful (World Economic Forum, 2007). This does not cover awareness raising
programmes and literacy related initiatives.
In addition to that, this comparison also uses expert estimations of the success of their governments in the
promotion of ICTs. Even if the scale is very subjective and based on qualitative overviews, it still gives some
estimation of the relative prominence of government activities.
Classification is on the scale where:
0 = none – there are no activities or activities are not related to concrete issues
1 = one, a little – there is one activity or a small-scale set of activities
2 = several, large scale – there are several initiatives or one complex measure addressing several aspects of
an issue.
The classification below is subjective based on reading of the materials provided by the EU Kids Online
partners. The scale and impact of the initiatives are difficult to estimate based on the short descriptions
available and thus these were also subsequently reviewed by each partner. However, the estimations match
relatively well with the NRI index. The differences may also arise from the fact that in many cases the
promotion of internet use has been more active in the past rather than currently.
32
Author: Pille Pruulmann-Vengerfeldt.
87
Table 3.8: Country variation in the ICT promotion and awareness raising
Austria
Belgium
Bulgaria
Cyprus
Czech Republic
Denmark
Estonia
France
Germany
Greece
Iceland
Ireland
Italy
Netherlands
Norway
Poland
Portugal
Slovenia
Spain
Sweden
UK
NRI result
on scale of
1-7
4,91
4,26
3,61
4,06
3,58
5,36
5,57
4,86
4,64
3,65
5,21
4,50
3,67
4,58
4,95
3,04
5,14
4,25
3,80
5,41
4,44
Use of the
Internet?
1
2
1
2
NA
1
2
2
2
2
2
2
1
2
2
1
2
2
2
1
2
Raise awareness of
potential social
impacts and risks?
2
2
1
0
Promote
media
literacy?
2
2
0
0
1
1
0
1
2
2
2
1
2
2
2
1
2
1
2
2
1
1
0
1
1
2
1
1
2
2
0
0
2
1
2
1
In general, one can say that it seems that where the internet is less common, then more efforts are made in
promotion of Internet use, while once the internet becomes more common, risk awareness and then literacy
initiatives become more visible.
These countries do not differ very much in terms of how much effort is spent on popularising ICTs – with
some exceptions (Bulgaria, Italy and Cyprus), where attention is mostly on ICT usage in schools and less on
use by private individuals. This seems also to correlate with level of internet usage in general.
In terms of internet safety, France and Bulgaria stand out as there are no significant internet safety
initiatives, Estonia and Cyprus seem to have few safety related initiatives. In terms of media literacy
initiatives, most of EU Kids countries seem to have just few initiatives. UK, Slovenia, Netherlands, Norway
and Austria stand out as being more active in terms of media literacy, while the reports on Germany and
France have no explicit mention of media literacy related initiatives.
In terms of activities related to promoting Internet use, the NRI sub-index provides good and comparable
data. Although being based on subjective opinions of experts, it seems to provide good overview. At the
same time awareness raising programmes and literacy programmes could be easily rated by partner
organisations in terms of the number of initiatives. However, some estimate should be made as regards their
reach. At the moment, the table composed above does gives only a limited comparative overview and the
styles of those writing the national reports differed, making the summaries less comparable.
One hypothesis would be that where governments are more active in promoting safety issues, there will be
more awareness of them. Here again the lack of empirical data on awareness of safety issues does not allow
to evaluate this hypothesis and to learn about efficient ways to promote safer internet.
88
3.2.3
The influence of NGOs
33
There are no comparable data.
When thinking about the influence of NGOs in the area of internet safety it is first worthwhile noting the long
tradition of NGOs working on general welfare issues related to children. For organizations which have
worked in areas of child protection and welfare, internet safety is to some extent a natural issue to take
onboard alongside other issues such as child poverty and violence against children. Out of 14 EU Kids
network countries nine report that their NGOs as active in lobbying both government and ISPs to impose
more regulation and stricter control on the internet in order to improve children’s safety online. These
countries are Austria, Cyprus, Denmark, Iceland, Ireland, Italy, Portugal, Spain and the United Kingdom.
Where NGOs are reported to be active in terms of lobbying they invariably seem to target both the
government and the ISPs. In five of these countries Austria, Iceland, Ireland, Italy and the United Kingdom
the NGOs are also reported to have been successful in influencing legislation or regulation. Moreover, in
three of these countries - Austria, Iceland and Italy - the national safety awareness node is run by an NGO.
Hence, it is clear that there are different levels of involvement in EU Kids Online countries ranging from very
limited to substantial. At one end of the continuum there are countries like in Bulgaria where there are very
few examples of NGOs trying to attract the public attention and where the atmosphere in general is not
favourable towards regulation. At the other end of the continuum there are countries like the five countries
mentioned above where there has been substantial pressure from NGOs towards increased regulation.
Based on the country reports it seems meaningful to distinguish between three groups of countries when it
comes to the influence of NGOs on legislation and regulation.
Table 3.9: Country variation in the influence of NGOs on legislation and regulation
Group one:
NGOs active and influential
Group two:
NGOs active but not influential
Group three:
NGOs not active
Austria
Denmark
Germany
Iceland
Ireland
Italy
United Kingdom
Cyprus
Portugal
Spain
Bulgaria
Czech Republic
Estonia
France
Greece
Slovenia
Sweden
There was incomplete information to perform classification for Belgium, Netherlands, Norway and Poland.
Information about how the national awareness nodes are organised combined with the above might provide
some further insight into the different cultures at play in the countries. For example, as already noted, in
three of the countries where NGOs have actively lobbied for stricter controls and regulation they are also
running the national awareness node. Another model is where NGOs are influential but the government runs
the awareness node either under the umbrella of the media regulating authorities (as in Denmark, Norway,
Sweden, Ireland and Germany) or even connected to the police (as in the UK). A third model is where the
government takes the initiative to create some kind of partnership to run the awareness node (as in Cyprus
and Portugal). A fourth model is to place the awareness node in the hands of a private company (as in the
Czech Republic and in Greece). Then there are interesting exceptions such as Bulgaria and Estonia where
there are no official awareness nodes and Slovenia where the awareness node seems to be an academic
initiative.
There may be more safety awareness in countries where NGOs are active, and more still in ones where they
are active and influential.
33
Author: Kjartan Olafsson
89
3.3 Public discourses
3.3.1
Media coverage
Since there were no data available to compare media coverage of children and the internet, some EU Kids
34
Online national teams conducted a 2 month analysis of selected national newspapers. This project asked a
range of questions about the overall tone of the articles (positive or negative), the parts of the internet
discussed, the source of the article, whose voice was heard, etc, but the main results discussed here relate
to risk. The figures reported are at best broadly indicative since there are some methodological challenges in
comparing the very different types of press in these different countries. Risk was analysed using the EU Kids
Online grid classifying concern about content, contact and conduct, and also commercial interests,
aggression, sexuality or values/ideology.
One thing that is clear is how varied the media coverage can be in the different countries, both in the
statistics and in studies of press coverage of single incidents, reported below. However, grouping countries
remains a problem. One clear observation is that common ways to cluster countries do not account for the
particular media coverage. In the tables and observations below, we have countries from north and south
Europe that are high or low by some criterion. The same is true for internet penetration e.g. the UK and
Denmark are in the same group as Portugal and Greece at one point. Media coverage must be driven by
other factors. One possibility, beyond what could be achieved in this simple content analysis, is whether
there are common patterns of conceptions of childhood that lie behind and are embedded in particular
national media coverage. For example, in Norway there is a notion of a ‘natural childhood’, where sexuality is
less of a risk while at the same time discussions of children’s rights is strong. Such underlying conceptions
may well help to shape the nature of how media engage in the topic of children and the Internet.
In all the countries in the table below what was common was the newsworthiness of risks compared to
opportunities – in all countries over half of all articles reported solely risks, the average of all these countries
being 64%, i.e. nearly two-thirds. In contrast, at most only a quarter of the media articles covered solely
35
opportunities in any country (i.e. they were all 25% of less) and the average was 18%. Looking in more
detail, the most important reason why risks predominate is that the chief source of news in most countries is
reporting on crime, mainly related to court cases and police actions.
Within the pattern outlined above there were some outliers:
The countries with a general high level of risk reporting in the media were the Portugal (85% of all articles),
36
the UK (77%) and (at face value ) Denmark (76%)
The countries with low levels of opportunity reports were Greece (5%), Portugal, (8%) and the UK (7%)
37
If we then look at the types of risk reported according to the EU Kids Online typology was see the following in
table 3.10:
34
See the table below. Bulgaria was also involved but the very low numbers of articles in the sample, reflecting little
coverage in that country, mean that it’s results are not included here
35
The remaining articles, usually a few percent in each country, covered a mixture of the risks we identified plus
opportunities, addiction or the dealt with something else besides risks
36
The problem is that Denmark, after Bulgaria, had the least number of articles reporting children and the Internet at all –
only 21 articles in the two months. This means that its percentages could be more easily influenced by just a few articles.
In addition, the Danish member of EU Kids Online pointed out that at least in some of the articles, when risk was
discussed it was not so much as a concern but there was a more reflective discussion of whether it should be a concern,
more neutral in tone. This qualitative consideration has to make us cautious in interpreting some of the quantitative data.
37
The reason this is not the same list as high risk list above is the same as footnote 2 – a few other percent are taken up
by these other categories.
90
Table 3.10: Country variation in the types of risks coded in relation to the three risk codes in the
38
national samples of articles
Risk/
Content Contact Conduct
Total
N
Country
25%
10%
65%
100%
59
Austria
55%
28%
17%
100%
94
Belgium
40%
44%
16%
100%
25
Denmark
54%
12%
34%
100%
158
Estonia
44%
13%
43%
100%
118
Germany
64%
23%
13%
100%
44
Greece
57%
16%
27%
100%
55
Ireland
29%
23%
48%
100%
90
Italy
22%
12%
66%
100%
79
Norway
Portugal
59%
23%
18%
100%
71
Slovenia
41%
34%
25%
100%
111
Spain
60%
13%
27%
100%
130
UK
54%
16%
30%
100%
50
Average
47%
21%
32%
100%
Media analysis carried out by EU Kids Online
Based on the table 3.10, we can see the following country classification in table 3.11:
Table 3.11: Countries ordered by whether media coverage is high
Risk
Content
Contact
reported
High
Low
High
Low
Level of
coverage
Countries
Greece
Spain
Portugal
Ireland
Norway
Austria
Italy
40
Denmark
Slovenia
Austria
Estonia
Germany
Spain
39
or low for different types of risk
Conduct
High
Low
Norway
Austria
Italy
Greece
Denmark
Belgium
Arguably the most striking point is that – at face value - different national media have very varied levels of
coverage of the three types of risk. Countries low on content risks like Italy, can be high on conduct risks,
and vice versa if we look at Denmark for conduct vs. contact. Or some countries can be high or low for some
risks, but be medium for others (in which case, they do not appear in the columns of this table). Hence,
media coverage in different countries is sensitising people to different kinds of risk, which may have a
bearing on how the degree to which people in different countries think the various risks are prevalent.
Let us look a little deeper at one type of content: that relating to sexuality, which is mainly coverage of
pornography on the net. The table is not reproduced here but there was a high interest in sexuality in content
in Belgium (42% of all items in the 12 cells of the grid), Greece (39%), Spain (37%), and the UK (36%). In
contrast, interest in this issue is shown to be very low in Norway (6%), Estonia (12%) and Denmark (12%).
Apart from the influence of particular national histories (e.g. the paedophile cases in Belgium), this probably
reflects different national concerns (at least in the media) about what images of sexuality children should be
exposed to. This in turn relates to national conceptions of childhood, as illustrated above in relation to
Norway.
38
Some articles were multi-coded – e.g. they might include content and contact elements. Therefore N is not the number
of articles, it is the total number of codes referring to the combination of content, contact and conduct in all articles in that
country. Content percentages are the number of codes referring to content divided by N, the total number of codes.
Hence, 25% of all codes referring to these 3 risks in Austria referred to content.
39
Obviously this can be achieved in a number of ways. The criteria of above and below average World produce a longer
list of countries. The table grouping by more extreme scores produces few countries that might make hypotheses texting
easier.
40
Same caveat as in footnote 2 – the Danish figures are based on fewer articles and hence fewer codes.
91
Finally it is worthwhile looking behind the figures. During the data collection period there were two
international stories which may have helped to shape the above figures, and we can take one to of these
cases to illustrate this process. A Finnish schoolboy killed his peers and teachers at a school and reported
his intentions on the Internet. There were also some subsequent related and copycat attempts in different
countries. All countries covered this but they did so to different extents.
The Norwegian, Austrian and Italian figures for ‘conduct risks’ are high because they had far more coverage
of this one story and subsequent events. This not only influences that particular column but all the other ones
– since such a high percentage were about conduct, a lower percentage in those countries were concerned
the other risks. So on the one hand the figures are ‘accurate’ in the sense that this was the coverage in the
time period. But are they ‘normal’ in the sense that would they have been different if this event had not
occurred? In the case of Norway, part of the reason for reporting was that it happened in a neighbouring
country and in recent years there had been public discourses about ‘looking to Finnish schools’ because the
Finns were performing better than the Norwegians in league tables. In addition one of the copycat attempts
was in Norway. All this would make the Norwegian coverage more understandable, and we might speculate
that coverage would have been less had the original incident occurred in a different country. But this would
not explain degree of Austrian and Italian reporting. Moreover Estonia also reported the case extensively, but
still did not appear high on the conduct criterion.
If we now look at content risks in the media, Portugal and Greece come out high. But both countries (along
with Austria) were amongst those with a high proportion of international news stories in general, (including
the second international story of a paedophile’s images on children on the Internet - this one story boosted
the ‘contents’ statistics). So the national press may cover risk stories, but that does not necessarily mean
‘risk in my country’. So the implications might be different from countries were national stories of risk
predominate.
In sum, we have a result, but it is almost a starting point for asking how one could take the interpretation of
media analysis further. For example, future research might look to see if media analysis of one particular
event can reveal culturally/nationally specific characteristics about media coverage/media attention and
public discourses.
Parents in the countries with a general high level of risk reporting in the media (Portugal, the UK and
Denmark – although see footnotes about the latter), will have a higher perception of risks than the average of
all these countries.
In countries where press coverage reports considerable concerns about the risks of content online, there will
be more parental concern about this issues compared to countries where that particular reporting is low.
The same logic applies to contact and conduct risks.
In countries where the media coverage reports considerable concerns about the risks of content online,
children will be more awareness of the issues compared to countries where that particular reporting is low.
The same logic applies to contact and conduct risks.
In addition, and not directly derived from the above observations, it may be assumed that the number,
strength and vocalness of agents that address the issue of children and online risks (and opportunities) in a
country will have an influence on national media coverage as well as parental concerns.
3.3.2
Role of NGOs and related stakeholders in shaping public discourses
41
In addition (and partly overlapping) to the above section on NGO’s influence on ICT regulation and related
politics this section examines to what extent NGOs influence the public discourses on online related risks
and opportunities.
There is material in the EU Kids national reports about the role of NGOs in this respect, with data missing
only for the Czech Republic and Poland. There are very few data about the activities of French NGOs in the
41
Author: Jivka Marinova.
92
field. However, different national teams have different approaches to the topic and one problem is that in
some of the reports the responses strictly follow the questions, in others there is a more narrative response.
In most of the countries participating in the EU Kids project NGOs have been playing an important role in
shaping both the policy and the public opinion. For example, the NGOs in Austria, Belgium, Denmark and
Estonia were particularly influential. In Germany NGOs have had many joint initiatives with companies
providing internet access and with universities. In Iceland, NGOs speaking with one voice were able to
shape the state policy. In Ireland the government already recognizes the child protection practitioners as key
stakeholders in the regulation of the Internet. In Italy, NGOs have shaped the legislative tools and framework
and the public discourse. In the Netherlands, thanks to the NGOs legislation on grooming was adopted. In
Norway also, very strong lobbyists promote parental awareness and legislation against grooming.
In Slovenia the Youth council has been influential in shaping the policy of the Office of the Republic of
Slovenia for Youth. In Spain the large NGOs working in the field have been instrumental in co-operating with
companies providing internet access. In the UK NGOs have been working with media and have been
supplying them regularly with results of different studies. There are some countries where NGOs are not very
influential. These are Bulgaria, Cyprus, France, Greece, Portugal and Sweden. However in both Bulgaria
and Cyprus as well as in Greece there are awareness nodes and hotlines which are trying to compensate
and also NGOs working in the field are getting stronger. In Portugal, although NGOs have a weak influence
they managed to attract public attention to the risks of the internet for children. Sweden is out of both
categories of countries. Although NGOs do not have big influence there is a strong governmental policy and
reliance of the public on State regulation for protection.
The main commonalities consist in target groups: in almost all countries NGOs are focusing on raising the
awareness of parents and children and to a lesser extent they target the service providers. Another
commonality is that very few NGOs deal only with safer internet issues. Most of those working on this topic
are NGOs working closely with national child protection agencies and more generally consist of child
protection organisations and some extent parents’ organisations as well. Mainly NGOs are running the
awareness nodes and the help lines.
Since the accounts provided in the reports are very uneven - some being detailed, some providing an
overview –, it is difficult to define criteria for classifying countries. Potential meaningful classifications would
be countries with and countries without awareness nodes and helplines/hotlines.
One hypothesis would be that in countries without awareness nodes, safety awareness would be lower.
3.3.3 Key Events42
In some respects this section provided a chance to discuss media coverage more generally, given that there
was no place to volunteer comments under the media heading because this is being handled by a separate
project. Hence the Czech, Greek and Irish contributions reflected critically on the national media in general,
while the Estonian reported details of the different types of media stories covered there.
In terms of events, it is easiest to look at what is common. In many countries there seems to be ongoing
media coverage of crime relating to children and the Internet, especially those concerning paedophiles.
Some nation teams reported this generally (Austria, Cyprus, Estonia, France, Norway, Portugal and the UK),
while others gave several particular examples (Greece and Ireland and Poland). It is clear from the media
analysis project that even if the national teams did not mention crimes under ‘key events’, reporting of such
crimes nevertheless takes place in their media (e.g. Spain and Bulgaria). In other words, the type of ‘event’
42
Author: Leslie Haddon.
93
that helps to maintain awareness in this area is the ongoing reporting of crimes. Sometimes it is the police
operations themselves (e.g. Austria and Poland) that gain visibility for this field.
In some case we have examples where particular high profile ‘crimes’ or ‘anti-social behaviour’ have
generated public discussions, as in the school killings in Germany and Finland – the latter generating a
major set of debates in Estonia (as was also clear from the media analysis). In Slovenia a particular positing
of youth misbehaving in school was the key event, while in France and Italy it was (separate) cases of
happy-slapping. So in addition to more general reporting of crimes, these specific ‘crimes’ or ‘anti-social
behaviour’ can also be very salient. The Portuguese case of the disappearance of the McCann girl shows
how a crime that in itself did not involve the Internet can nevertheless raise wider debates about children
online. A more general discussion of the Internet also emerged in Estonia after the Finnish school massacre,
which once again shows the power of crimes to frame and evoke discussions
In a few countries it was events such as conferences (Bulgaria, Greece), projects (Estonia, Norway),
campaigns (Iceland) and even the Safer Internet Day itself (Bulgaria, the Netherlands, Spain) that were
reported. In the majority of countries, such events clearly did not immediately come to mind as ‘key events’.
The differences across countries lie in relation to the influence of ‘positive’ events, and certainly there is no
clear pattern as regards geography (e.g. North-South/East-West, GDP or the length of time that the Internet
has been established). At best you could say that teams from the larger countries (in terms of population) did
not mention such positive events.
From a more general knowledge of the national reports and past comments about Safer Internet Day in
particular, it looks as if it is more common for positive initiatives such as conferences, projects, campaigns
and Safer Internet Day to be influential in those countries and at that point in time when there was previously
little awareness of risks. For example, nowadays Safer Internet Day gets little media coverage in the UK, and
the Norwegians have pointed out the same is true there given that the issues have by now been in the public
eye for a few years.
3.4
3.4.1
Values and attitudes43
Sources of information
The European Values Survey (EVS) provides a reliable pan-European source of data on adults’ values,
informed by Inglehart’s analysis of value orientations (Inglehart and Baker, 2000). This analysis has been
used to characterise countries according to their relative endorsement of individualistic and collectivist
values. Thus, people in individualistic cultures engage in open interpersonal emotional expression in order to
attain their personal well-being, and have a need for autonomy, independence and individuality. Collectivism
has been conceptualized as a worldview in which group membership occupies a central place and the
importance of self is only peripheral. Hence people in collectivist cultures restrain their personal emotional
expressions and emphasise obedience and unselfishness in relationships.
3.4.2
Commonalities and differences between the countries
The classification of countries is based on adults’ answers to eleven values measured in the EVS: the
importance of good manners, independence, hard work, responsibility, imagination, tolerance and respect,
thrift and saving money, determination and perseverance, religious faith, unselfishness, and obedience.
While the differences are a matter of degree, a classification of countries is possible on this basis.
3.4.3
Indicators for classifications of the European countries
Using the EVS data from 2000, Kirwil (2008) conducted a factor analysis in order to classify the EU Kids
Online countries according to their value orientations. Two independent factors were identified, which can be
43
Author: Lucyna Kirwil
94
interpreted as individualistic and collectivistic orientation. As table 3.12 shows, the countries differ quite
substantially with regard to these two dimensions.
Table 3.12: Country variation in value orientation
Individualistic
Values Orientation
,669
,344
-1,195
Data not available
-,609
1,757
-1,208
-,091
,456
-,130
,364
,538
,171
,987
Data not available
-1,429
-,840
,629
,416
1,403
,998
Austria
Belgium
Bulgaria
Cyprus
Czech Republic
Denmark
Estonia
France
Germany
Greece
Iceland
Ireland
Italy
Netherlands
Norway
Poland
Portugal
Slovenia
Spain
Sweden
UK
Collectivistic Values Orientation
-1,035
,183
,227
Data not available
-,087
-1,030
,191
-,040
-,983
-,070
-,580
,274
-,140
-,899
Data not available
,470
,339
-,264
-,672
-1,162
,534
On the basis of the countries’ values on these two factors four clusters were identified:
1.
High/Moderate Individualism and Moderate Collectivism: UK, Ireland, Belgium
2.
Low individualism and Moderate Collectivism: Poland, Bulgaria, Estonia, Portugal, and Czech Republic
3.
Moderate Individualism and Low Collectivism: Austria, Germany, Slovenia, Spain, Iceland, Italy, France
and Greece
4.
High Individualism and Low Collectivism: Denmark, Sweden, Netherlands.
To a great extent, the groups obtained follow the structure proposed by Inglehart and Baker (ibid) for:
Group 1: English speaking countries plus Belgium
Group 2: Ex-communist countries, though Portugal is also included here
Group 3: Catholic Europe
Group 4: Protestant Europe
The group values on the two dimensions are shown in the table below (table 3.13).
Table 3.13: Country clusters according to value orientation
Values
Dimension
Individualism
Collectivism
Cluster 1
UK,
Ireland,
Belgium
,67
,36
Cluster 2
Bulgaria, Czech
Republic, Estonia,
Poland, Portugal
-1,06
,23
Cluster 3
Austria, France,
Germany, Greece,
Iceland, Italy, Slovenia,
Spain
,33
-,52
95
Cluster 4
Denmark,
Netherlands,
Sweden
1,38
-1,03
Total
Sample of
17 European
Countries
,15
-,29
3.4.4
Hypotheses regarding the influence of this contextual factor on Safer Internet
It seems plausible that the value orientation typical of a country might influence parental strategies in
mediating their children’s internet use. For example, monitoring the child is one of the most accepted
parental techniques in socialisation of a child, and assumes that the child is not directly controlled by the
parent. Hence it may be hypothesised that this approach would be preferred by parents with an
individualistic values orientation. Parents from collectivistic countries may be expected to mediate the
internet in a more direct, interactive way than parents in individualistic countries, who may prefer more
efficient but indirect parenting (e.g. abstract rule setting). At the same time, children from collective cultures
should maintain harmonious relationships by not contradicting parents when compared to children from
individualist cultures. Therefore direct interpersonal parental mediating of the Internet should be more
efficient in collectivistic countries than in individualistic countries.
Other hypotheses could be formulated. Given the widely acknowledged importance of cultural values,
including as a potential influence on parenting and socialisation practices, these country classifications are
surely worth pursuing in further research.
3.5
3.5.1
Educational system44
General literacy of the population
OECD data is available showing the educational attainment of the adult population.
General literacy rates in most of the 21 European countries are generally high. From northern to southern
Europe, from Eastern to Western Europe, all countries show a literacy rate of at least 90%.
Nevertheless, in some countries this optimistic scenario is contested. On one hand, official statistics are not
without inaccuracies, thus being disputed by alternative data (e.g. in the UK). On the other hand, functional
illiteracy is usually unnoticed in most reports based on official data. Some countries (e.g. Ireland, UK and
Belgium) have clearly pointed out this problem.
Although the basic level of literacy and primary education is found in most countries, secondary education,
and even more so higher education, is less frequent (HE is 10% to 20%, depending on country). This is why
for some education may still be considered more “elitist” than “mass”. Nevertheless, most countries reported
considerable growth participation in higher education.
44
Author: José Alberto Simoes.
96
Table 3.14: Country variation in educational attainment: adult population (2005)
25-64 year old by highest level of education attained (%)
Country
Pre-primary
and primary
education
Secondary
education
Lower
Upper
19
54
18
33
Post-secondary
non-tertiary
education
Austria
9
Belgium
15
2
n.a.
n.a.
n.a.
n.a.
Bulgaria
n.a.
n.a.
n.a.
n.a.
Cyprus
Czech Republic
10
77
Denmark
1
16
50
Estonia
1
10
49
7
France
14
19
42
Germany
3
14
52
6
Greece
29
11
32
7
Iceland
3
28
36
3
Ireland
17
18
25
11
Italy
17
32
37
1
Netherlands
8
21
38
3
Norway
22
41
4
Poland
15
65
4
Portugal
59
15
13
1
Slovenia
2
17
60
Spain
24
27
20
Sweden
7
10
48
6
United Kingdom
14
56
Source: adapted from OECD, Education at a glance (OECD, 2007).
n.a. Not available.
Tertiary
education
18
30
n.a.
n.a.
13
34
34
25
25
21
31
29
13
31
33
17
21
20
28
30
30
If we look closer at the educational level achieved by the adult population for each country (table 3.14) the
above scenario becomes less bright. There are clear differences between countries, just as there are some
similarities.
!
Southern Europe countries, like Portugal, Greece and Spain show considerable high rates of only preprimary and primary education as compared to other countries (59%, 29% and 24%, respectively). The
case of Portugal is particularly outstanding, given the fact that according to the OECD data (OECD,
2007) more than half of its adult population never got beyond primary level of education.
!
In contrast, in Northern European countries (Norway, Denmark, Iceland and the UK), Eastern European
countries (Czech Republic, Poland and Estonia) and central Europe countries (Germany and Slovenia),
only a small proportion (less than 3%) of their adult population has never achieved more than primary
education.
!
Ireland, Italy, Belgium, France and the Netherlands, scattered from north to south of Europe, remain in
an in-between position.
Participation in tertiary education generally reaffirms the previous interpretation. The countries that show
lower rates of only pre-primary and primary education are the ones that, conversely, show higher rates of
participation in tertiary education.
!
Apart from Belgium (central Europe) and Estonia (north-eastern Europe), we should note that all the
countries that stand out as being above average (25.4%) are located in northern Europe: Denmark,
Norway, Iceland, Netherlands, Sweden and UK (all with 30% or more). Ireland remains slightly below
that figure (29%).
The above rule comes, however, with some exceptions:
!
Despite the fact that Portugal presents the highest rate of people with only the basic level of education,
as regards tertiary education the country is not at the bottom of the scale (21%). We might say the same
(with a smaller discrepancy) of Spain’s tertiary achievement rate (28%).
97
!
Other countries remain exceptions but for the opposite reason: their tertiary rate is far lower than would
be expected given their attainment beyond primary education. This applies to the Czech Republic, Italy
(both 13%), Poland (17%) and Austria (18%). Secondary education, however, seems to have great
importance in those countries.
The above tendencies are noticed but not really explained. Thus we have to take into consideration other
social indicators and especially elaborate on contextual factors that might clarify the pattern described. In the
following sections we will try to introduce some comments as well as other figures that might help us
accomplish this task.
One hypothesis would be that there would be lower safety awareness in countries where the educational
level was lower – especially where a higher proportion experience only primary education. Beyond that a low
level of education might result in many parents not being able to help their children when using the internet.
For the specific case of Portugal it has been shown in chapter 2.1.1 that Portuguese adults are very unlikely
to use the internet and that one fifth of the young internet users in Portugal uses online media while their
parents do not (see table 2.1).
3.5.2
The education of the parents’ generation
OECD data is available showing the educational attainment of the adult population.
All social activities are a product of particular historical circumstances. Education is no exception.
Generational differences in educational patterns indicate more general social transformations that have
occurred in each country. That is particularly the case of countries that have experienced a considerable
expansion of their educational systems in the last decades.
In the section on literacy and general education we examined educational attainment by level of education.
This gave us some idea of general levels of education at a given point in time. In this section we would like to
consider generational differences in order to compare the parents’ generation with their children’s present
45
experience . Since we do not really have longitudinal data referring to exactly the same children studied at a
particular moment, we have to build our interpretation of change on the basis of age differences examined at
a given moment (which might tell us something retrospectively about each generation).
In order to simplify this analysis we will focus on tertiary education. More specifically we will be considering
tertiary level of education attainment by age group in each country (table 3.15) as a specific indicator of the
educational achievement of diverse generations.
45
Our chronological references are somewhat vague on this point. Each country reported their generational changes
with specific data sources and criteria, thus making comparisons difficult. In fact, to answer truthfully the above question
we would have to consider the following problem: since children from the same generation (i.e. born approximately in the
same period of time) might have parents with different ages (and therefore from different generations), how do we
establish which generations to compare? To be more exact, how do we measure “parent’s generation”? At what
date/period of time do we start? This is the first decision to make. Another option would be to consider the education
level of the parents of children/ young people surveyed at a given moment in time.
98
Table 3.15: Country variation in the proportion of the population that has attained tertiary education
by age group (2005) (%)
Age groups
Country
Ratio of the variation
between the youngest and
(1)
the oldest age group
25-34
35-44
45-54
55-64
25-64
Austria
20
19
17
14
18
Belgium
41
33
27
14
31
n.a.
n.a.
n.a.
n.a.
n.a.
Bulgaria
n.a.
n.a.
n.a.
n.a.
n.a.
Cyprus
Czech Republic
14
14
13
11
13
Denmark
40
35
32
27
34
Estonia
33
36
35
29
33
France
39
25
18
16
25
Germany
22
26
26
23
25
Greece
25
26
19
12
21
Iceland
36
34
29
21
31
Ireland
41
30
22
17
29
Italy
16
13
11
8
12
Netherlands
35
30
30
24
30
Norway
41
35
30
24
33
Poland
26
16
12
13
17
Portugal
19
13
10
7
13
Slovenia
25
21
17
16
20
Spain
40
30
22
14
30
Sweden
37
28
28
25
30
United Kingdom
35
30
28
24
30
Source: adapted from OECD, Education at a glance (OECD, 2007).
n.a.
Not available.
(1)
Difference between the age group 25-35 and the age group 55-64 divided by 55-64. Not available
0,4
1,9
n.a.
n.a.
0,3
0,5
0,1
1,4
0,0
1,1
0,7
1,4
1,0
0,5
0,7
1,0
1,7
0,6
1,9
0,5
0,5
on the original table.
Except for Germany, in all countries the older one gets the less one is expected to attain a tertiary education.
This tendency is important not because it reveals age differences but because it gives you an idea about
generational differentiation. Younger people are more likely to proceed to tertiary education than their
parents did. In fact, almost all percentages decrease linearly from younger age groups to older ones. This is
quite obvious in some countries, since the distance in educational achievement between younger and older
age groups is rather accentuated. Even though virtually all countries show the same tendency, there are
considerable divergences among them in what concerns the degree to which this is discernible.
!
Spain, Belgium, Portugal, France and Ireland are amongst the countries with the highest variation
between age groups. Spain and Belgium have almost twice the proportion of people in the youngest
age group in tertiary education compared to the older group. As for the other countries, they all have at
least one and half times the proportion of 25-35 years old as compared to 55-64 years old.
!
In the case of Greece, Italy and Poland the proportion 25-34 years old is at least once more than the
proportion found on the oldest group.
!
Estonia, Czech Republic, Austria, the UK, Sweden. Denmark, the Netherlands and Slovenia, they all
show smaller differences between the oldest and youngest age groups, in some cases practically
unnoticeable (e.g. Estonia).
By taking tertiary education as an indicator of the transformations undergone by the educational system of
each country and most of all as an indicator of generational differentiation in learning opportunities, it has
become apparent that there are dissimilar speeds and degrees of change. Some countries started their shift
toward the generalisation of schooling and of higher education in particular at an earlier stage, while in
others this is a relatively recent process.
However, the expansion of the educational system and in particular of higher education between generations
is not synonymous with “democratisation”. For instance, the expansion of higher education does not mean
46
that everyone has the same chance of entering the system . So, social inequalities tend to reproduce
46
And even if they do, they may get into less prestigious schools or courses.
99
themselves not only with regard to accessing higher levels of education, but also in the internal differentiation
within the system itself. This has been noted by teams in countries such as Greece and Portugal in their
country reports.
The Eurostudent report of 2005 (Eurostudent Report, 2005) provides us with some indicators that are helpful
for understanding socio-economic differences, even though only a small group of European countries (10) is
considered. These indicators examine differences in the socio-economic status of higher education students
by their fathers’ educational background. “In many countries, students are substantially more likely to be in
higher education if their fathers completed higher education. Students from such a background are more
than twice as likely to be in higher education in Austria, France, Germany, Portugal and the United Kingdom
than are students whose fathers did not complete higher education. In Ireland and Spain this ratio drops to
1.1 and 1.5, respectively.”
According to the same report, “inequalities in previous schooling are reflected in the intake of students from
less advantaged backgrounds. Countries providing more equitable access to higher education – such as
Finland, Ireland and Spain – were also the countries with the most equal between-school performances in
47
PISA 2000” (Eurostudent Report, 2005).
Consequently, if on the one hand the educational system contributes to engender expectations regarding
equal opportunities and social mobility, on the other hand it appears that the system is functioning as a
48
ground for cultural reproduction and the perpetuation of social inequalities . So the question is how
accessing the school system, in all its stages, might reflect and reinforce social inequalities.
3.5.2.4 Hypotheses regarding the influence of the contextual factors on Safer Internet issues
One hypothesis is that there would be greater safety awareness in countries where the older/parents’
generation had a higher level of education.
The differences with regard to the role of parents’ education and social background for the education of their
children might explain some of the (unsystematic) observations concerning the role of socio-economic status
on children’s online opportunities and risks.
3.5.3
The kind of education for today’s children
The following overview is based on the EU Kids national reports.
One way or another, all countries seem to have experienced changes in their educational systems in the
past decades, both quantitatively and qualitatively. Some structural transformations took place, first of all
through the longer participation in schooling for bigger parts of the population, but also through changes in
curricula and educational methods. More subjects, both diversified and updated, are now offered to students,
consequently presenting more choices. Several innovations have been introduced, such as new pedagogies,
new learning tools and methods. New technological tools have been adopted (including the internet and
other ICTs), which ultimately lead to a sense that the educational system seem to have become more open
and dynamic. If these tendencies are to some extent general their adoption, they have occurred at different
paces in different countries. For instance, Nordic countries have experienced an earlier expansion of their
educational system, while in southern Europe this happened latter. Meanwhile, Eastern countries were
relatively “closed” until recently, while most of western countries have “opened up” long before.
From a subjective point of view, the level of expectations seems to be higher now than before. This is
explained in part because the investment in education has grown considerably at all levels, but also because
the role assigned to education seems to be more important than ever.
One of the most obvious transformations that educational systems have undergone is their growth. Not only
has the number of students taking part in compulsory education increased but also this has occurred at all
47
Programme for International Student Assessment.
This is the classical thesis, among other authors, of the French sociologist Pierre Bourdieu (Bourdieu and JeanClaude, 1964; Bourdieu and Jean-Claude, 1970).
48
100
49
other levels of education . However, there is a more general problem regarding not only the development of
all levels of education but their actual configuration and meaning. This has been formulated in discussions of
the degree to which may we talk about the “democratisation” of the educational system. Is today’s education
less “elitist” than it was before? Taking several of the indicators that have been examined as evidence the
answer should be affirmative. But even so this does not mean that, as discussed in the previous section,
social differentiation has disappeared completely. This is particularly the case of countries in which socioeconomical inequalities are evident, since this is one of the mechanisms that explain access to the system.
The discussion of whether a particular educational system is “elitist” or “mass” is more complex than appears
at first and may be considered at different levels of education. In the compulsory system, social inequalities
regarding access have mainly been discussed in terms of the right to attain a basic education (which has
been recognized legally in a minimum years of schooling). On the other hand, as regards tertiary education,
access itself is selective, not only because certain schools have a better reputation than others but also
because the system itself leaves a number of potential candidates behind. In this case we may say that the
system (or at least part of it) is intrinsically discriminatory. Nevertheless, even in countries where the
educational system is (or has been for quite some time) apparently more selective than in others social
differentiation is becoming somehow mitigated by the lengthening of schooling. This is true in all levels of
education but most of all in higher education.
In countries that have experienced in the past decades a shift from what might be called “closed” societies to
“open” ones, through more or less profound political, cultural and economical transformations, perceptions of
change in the educational system is somehow obvious. This is the case of former communist countries (e.g.
Bulgaria, Czech Republic, Estonia), but also in southern European countries deprived of democracy until the
mid 70’s, like Portugal and Spain. Besides these structural transformations there are other changes more
specific to the educational system which occurred in the last two to three decades that should be mentioned.
In spite of the particular configuration that all these transformations assume in each country there are
50
commonalities worth mentioning. We may sum up these changes in just a few key dimensions :
School programs and resources
! The general transformation of curricula (from primary to tertiary education), including
diversification of choices, new subjects, etc.
! Media and ICTs used for learning and as taught subject (both specific and common).
! New pedagogies (new teaching methods, learning objectives, etc.)
Organisation of schools and institutional change
! General modification in students and teachers roles
! Weakening of authority of schools as regards students’ behaviour
! Decrease of school control over students (both as source of discipline and in temrs of having a
monopoly on knowledge).
Number and type of schools offered
! Specialisation and diversification of schools (more or less at all levels of education)
! Emergence of private schools at all levels of education (thus introducing further differentiation in
to the system).
Expectations about the role of education
! More investment in children’s and young people’s education (perceptible, for instance, through
parents’ concern and engagement in their children’s education)
! Importance assigned to education as a way to insure one’s future (particularly evident if we
compare the children’s generation with their parents)
In spite of these modifications some countries have reported evident continuities, in some cases in a positive
way (e.g. Iceland), in other cases in a negative way (e.g. Greece). In the first case this is because some
countries start off from a more favourable position, differences between generations are less strong, since
some of the goals of greater participation have already been attained. In the second case this is precisely
because some of the above innovations are far from being fully accomplished. Pedagogical methods have
49
The number of students at each level of education, the minimum (compulsory) years of schooling and extended hours
spent in school are some of the indicators that might be considered to illustrate this point.
50
Besides differences by country it would be necessary to note specific consequences at each level of education.
101
not been altered in a revolutionary way. The same may be said about the way objectives are defined and
evaluation methods are applied. In some of these countries diversification of curricular choices is not seen as
an entirely good thing, since it may lead to fragmentation and dispersion. On the other hand, as noted
previously, class based differentiation (grounded on socio-economic differences) has not been completely
eradicated (and there is no reason to believe that will be in the near future), so education is far from fulfilling
the utopian society, based on knowledge and information, aspired to by some.
In sum, economical, political, cultural and social contexts still matter and may explain a lot. An adequate
explanation of the implications from the above observations is beyond our ambition in this overview. It is only
possible to point out some dimensions that might be important in order to understand these variations.
Although there are similarities (and differences) outlined above, it was not possible to construct a
classification system that would differentiate among the EU Kids Online countries.
Without a classification system, it was not possible to develop hypotheses
3.5.4
The technical infrastructure of schools
Eurydice, 2005 (based on OECD, PISA 2000 and 2003).
As noted earlier, one of the areas where educational systems have changed considerably in the past years
is in terms of investment in the technological infrastructure of schools. Government investment in ICTs has
grown considerably in all countries over the last few years. In some cases specific organisation were created
to implement the use of the internet and other technologies in schools (e.g. Becta in the UK).
In addition, several attempts have been made to promote computer literacy among children (see next
section). In fact, almost all countries mentioned the importance of ICTs, not only as tools for learning but also
as an area of concern regarding younger generations.
Even though almost every school is connected to the Internet in most countries, the number of computers
per student is not as high as we would have thought, even if we may note a clear variation according to the
level of education (generally in higher levels of education there are more computers per student).
According to available data (Eurydice, 2005), there are slightly differences if a child attends the public sector
or a private school. Computer facilities in private schools are better in countries were schools are largely
founded by tuition fees.
Considerable changes have been noted as regards the computerisation of schools in European Union
countries (e.g. Greece, Poland and Portugal) over the past few years. All these countries have achieved a
level of at least one computer for every 20 students. In the majority of EU countries, however, the current
51
level is one computer for every ten pupils – or even for every five pupils (Eurydice, 2005) .
The number of schools with internet access has also increased significantly (see table 3.16) Data from Pisa
survey 2003 (see ibid) reveal that on average at least 60% of schools have computers connected to the
Internet.
Looking at table 3.16, we may confirm a previous observation. All countries have experienced clear growth.
In some cases rate of change is quite astonishing: Italy has almost twice as much schools connected to the
Internet in 2003 than in 2000; Greece has one and half times more schools connected to the Internet; Poland
almost one and half times; the Czech Republic doubled its proportion. The only countries that do not show
51
Based on OECD data from Pisa 2000 and 2003 projects. For more information, go to
http://www.oecd.org/pages/0,3417,en_32252351_32236130_1_1_1_1_1,00.html. See Annex Table 3.a with data based
on each country report.
102
significant growth are the ones that already had high proportions of internet connection (e.g. Iceland and
Sweden).
Table 3.16: Country variation in the average proportion of computers with internet connection in
schools attended by students aged 15, public and private sectors combined
Country
2000
2003
Variation rate
*
Austria
69.3
87.3
26,0
**
**
Belgium
47.2/-/42.6
65.2/71.6/79.8
38.1/-/87.3
Bulgaria
28.5
Not available
Not available
Not available
Cyprus
Czech Republic
39.8
79.8
100.5
Denmark
65
87.8
35.1
Estonia
40.7
79.3
94.8
France
26.3
Germany
37.7
70.7
87.5
Greece
26.4
69.2
162.1
Iceland
82.6
95.7
15.9
Ireland
46.6
67.4
44.6
Italy
24.1
70.8
193.8
Netherlands
84.8
Norway
49.8
81.2
63.1
Poland
35.3
82.7
134.3
Portugal
35.3
60.4
71.1
Slovenia
Spain
40.7
79.3
94.8
Sweden
74.3
91.9
23.7
***
UK
53.8/-/30.9/37.8
Source: adapted from Eurydice, 2005 (based on OECD, PISA 2000 and 2003).
*
Not available on the original table.
**
Respectively: French Community/ German-speaking Community/ Flemish Community.
***
Respectively: England/ Wales/ Northern Ireland/ Scotland.
On the other hand, as several countries reports have pointed out, internet penetration is not the same as
actual use. Most students cannot use Internet at schools without some kind of control by adults. Even in
tertiary education access is not completely without restrictions. The importance of some kind of mediation
from the school system is particularly evident when it comes to internet use by children.
103
Table 3.17: Country variation shown by indicators of technological infrastructure of schools
Country
Internet penetration rate in
Ratio student/ computer year Internet use
in schools
schools (%) year
Primary Secondary Tertiary Primary Secondary Tertiary
Nearly
Austria
100
2004 (not specified by grade)
not available
6
Existence of specific
government organisation for
ICT implementation in
schools
not available
Belgium
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
not available
Yes (Federal Action Plan)
Bulgaria
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
not available
not available
Cyprus
n.a.
n.a.
n.a.
7
not available
not available
17
2006
Czech Republic
Denmark
Estonia
Greece
100
2006
2006
2006
2006
25
Use with
restrictions
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
not available
not available
No restrictions
not available
not available
2006
6-7
10
14
Nearly
Nearly
n.a.
n.a.
n.a.
n.a.
100
98
100
99
Use with
restrictions
n.a.
n.a.
n.a.
n.a.
not available
2006
2006
96
Nearly
Yes
Yes
(Action Plan eLearning/
German Confederation)
n.a.
n.a.
n.a.
Use with
restrictions
not available
n.a.
n.a.
n.a.
No restrictions
not available
7.1
n.a.
Use with
restrictions
Iceland
(Internet do !kol)
100
2006
100
Germany
2006
100
France
7.7
2006
100
no year mentioned (not specified by
grade)
Nearly
Ireland
100
Italy
86
Nearly
2005
10.9
recent
Norway
9.1
2005
Yes
(NCTE)
not available
not available
100
no year mentioned (not specified by
n.a.
n.a.
Use with
restrictions
not available
n.a.
n.a.
n.a.
n.a.
not available
not available
grade)
Poland
Portugal
n.a.
n.a.
100
100
2006
Slovenia
Spain
2006
n.a.
n.a.
100
95.6
Sweden
81.1
n.a.
Not specified
The
Netherlands
The United
Kingdom
n.a.
n.a.
n.a.
100
not available
12.8
2006-07 (not specified by grade)
9
Yes
(Crie)
n.a.
n.a.
not available
not available
2006
7.1
n.a.
not available
not available
2005
2005
n.a.
n.a.
n.a.
not available
not available
n.a.
n.a.
n.a.
not available
not available
7
3
n.a.
not available
10.1
Yes
(Becta)
Source: Based on national reports for EU Kids Online.
There are so many factors at work here (level of internet access in school, level of growth, restrictions on
use) that ‘simple’ hypotheses might be inappropriate.
3.5.5
Internet and media education
Logically, internet and media education should be complementary to the technological infrastructure of
schools. However, regardless of how obvious this may seem investment in the technological infrastructure of
schools is not always followed by corresponding investment in ICT education.
104
We should differentiate, however, general use of ICTs and other media as tools for learning from ICTs and
media as subjects on the official curricula. The first case includes informal ICT and other media learning,
while the second case is related to formal learning about these technologies, whether as specific subjects or
as cross-curricular subject (common to several courses).
Except for Bulgaria, Cyprus and Sweden, all other countries reported that ICT learning is part of the curricula
(both in primary and secondary levels of education). In most countries ICT learning constitutes an
autonomous subject. Only in a few countries it is just a cross-curricular subject.
Table 3.18: Country variation in ICT and Media Education in schools
Country
ICT learning in schools
curricula
Other initiatives regarding
ICT use
Media education in
curricula
Austria
Belgium
Yes
Yes, in both primary and
secondary levels
Yes, in all subjects
Not mentioned
Bulgaria
Cyprus
Czech
Republic
Denmark
No
No
Yes, in secondary level
Not mentioned
ICT Knowledge centres (for
students, teachers,
unemployed)
No
Cyber Ethics events
Not mentioned
Yes, in both primary and
secondary levels
Estonia
Yes, cross-curricular
France
Yes, in primary and secondary
levels (B2i -Brevet Informatique
et Internet, since 2000)
Yes (but technical)
Germany
Greece
Yes, in primary (since 2001)
Yes
Yes, cross-curricular in primary
level and specific subject in
secondary level
Yes, in primary and secondary
Italy
levels
Netherlands
Yes, cross-curricular in primary
Norway
level and specific subject in
secondary level
Poland
Yes
Portugal
Yes, cross-curricular in
Slovenia
secondary levels
Yes
Spain
No
Sweden
Yes
UK
Source: Based on national reports.
Iceland
Ireland
IT training of teachers, IT as
optional subjects for secondary
grade
ICT educational programmes
No
Not mentioned
Yes, in primary level
yes
Yes, courses in
media education and
upper secondary
Not mentioned
Not mentioned
Not mentioned
Yes, Greek School Network
Learning material from the
portal of MoE
Not mentioned
Not mentioned
No
Not mentioned
Yes, with ICT
Not mentioned
Not mentioned
Not mentioned
Not mentioned
Not mentioned
Not mentioned
Not mentioned
No
Not mentioned
Not mentioned
Not mentioned
Not mentioned
Not mentioned
Not specifically
Not mentioned
Media and Internet education in schools is connected mainly with government’s engagement in promoting
technological literacy. Nonetheless, ICTs’ presence in the curricula does not cover all official initiatives
concerning this matter. Other institutional programmes regarding ICT education are also mentioned by some
country teams (e.g. Belgium, Cyprus 52, Greece, Denmark and Estonia).
It is difficult to assess if media education is part of national curricula or even a real educational concern since
most country teams did not answer this question specifically. In fact, only six countries reported any specific
concern for media education in the curriculum (Austria, Czech Republic, Estonia and Iceland), which is not
conclusive.
Being able to know which countries have (or do not have) internet and media education in their schools
curricula is one thing. Knowing what type of education we are talking about is another. Learning how to use a
52
In this case this is particularly important since ICT learning is not part of the curriculum.
105
computer is not only a matter of expertise. It is also a matter of knowing the implications of different kinds of
usage. Some countries’ reports have shown concern about ICT education being too technical (e.g. Ireland
and Iceland) and not really concerned about risks and opportunities of the internet and other ICTs.
Although we have some indications of similarities and differences, there is incomplete information on which
to classify countries into groups.
In principle, in countries with more internet and media education one might expect more safety awareness,
but as noted above, that depends on what is taught under these headings.
3.6
Background factors53
3.6.1 Levels of social change
It is of course difficult to identify the most important or most substantial social change both nationally and
internationally. Certain themes are, however, recurrent in many national reports. The state of the economy is
one and many of the countries represented in the EU Kids network have experienced rapid economic growth
over the past decade or so. This applies for most of the eastern European countries but also for others - for
example, the UK, Spain, Portugal, Ireland and Iceland. Demographics is another area where many countries
have experienced change, with a substantial inflow of migrants. This is the case, for example, for Greece,
Iceland, Ireland and the UK.
Another significant demographic change is the diminishing fertility rate, as for example in Austria and
Bulgaria. Culture is mentioned in some national reports but not as the most substantial change. Cultural
change is an important theme in many former eastern European countries that have experienced the change
from communism to capitalism. This change is of course not merely an economic change. Linked to that is
the change in Portugal and in Spain from dictatorships to democracies.
Enthusiasm and support for anything related to the information society and the internet seems to be the
general rule for the EU Kids countries. Many governments have actively worked to strengthen infrastructure
and otherwise increase the use of information technologies. Discourses on the importance of information
technologies do not, however, always materialise in concrete actions. In fact, it is also noteworthy that there
are countries where this discourse is simply not very visible.
It is therefore possible to classify the countries into three different groups. In the first group are countries
where discussions associated with the information society are high profile and where there are also concrete
actions aimed at strengthening the country’s’ position, mostly through investment in infrastructure. In the
second group there are countries where there is a public discourse on the importance of information
technologies but where limited action is taken, for example, to improve access or facilitate development. In
the third group are the countries where there is limited or no discussion of the issue. Based on the
information given in the national reports the countries can be classified into these groups as follows:
53
Authors: Kjartan Olafsson, Katja Segers, Liza Tsaliki.
106
Table 3.19: Clusters of countries according to discourses about ICTs
Group one:
discourse and action
Austria
Belgium
Cyprus
Denmark
Estonia
France
Iceland
Ireland
Portugal
Slovenia
Spain
Sweden
UK
Source: Based on national reports
Group two:
just discourse
Group three:
limited or no discourse
Czech Republic
Germany
Greece
Italy
Bulgaria
There is a close relation between the discourse on the information society and the general feeling of how the
country is doing in comparison to other countries. Thus, where there is an open debate on the importance of
the information society there are mainly two kinds of countries. There are on the one hand those who think
that they are on the leading edge and on the other hand those who think they are not.
If we take the table above to be justifiable then taking the analysis one step further would yield the following
classifications:
Table 3.20: Clusters of countries by self-evaluation of whether they are on the leading edge in
relation to the information society
Group one:
Think they are on the leading edge
Group two:
Think they are not on the leading edge
Austria
Denmark
Estonia
Iceland
Sweden
UK
Belgium
Cyprus
France
Ireland
Portugal
Slovenia
Spain
One hypothesis is that there would be more safety awareness in countries where there is discourse and
activity. Another is that there would be more in leading edge countries. In all it is highly plausible to assume
that a societal context, that is shaped by a high motivation to support new technologies and the opinion to be
among the leading countries will further children’s and parents’ interest in the internet. The above
classifications partly correspond to the classification according to the percentage of young internet users.
3.6.2
Inequalities
Regular data on inequalities in today’s societies are provided by the OECD.
Despite strong economic growth in most European countries for more than a decade, all European countries
still experience strong inequalities. Poverty is experienced by around 10% or more of the European
population. Throughout the whole of Europe, key divides are about the same. Vulnerability to poverty and
experience of social exclusion is especially found in certain social groups: women, lone parents/mothers,
elderly and disabled people, people living in rural areas, people with a weaker social background, immigrants
and European ethnic minorities (e.g. the Roma in Czech Republic and Bulgaria). Being part of those social
groups minimises the chances of achieving a higher education and equality, despite the post-war expansion
of educational participation throughout Europe.
107
Strong regional differences can be noticed within politically regionalised countries such as Spain (where e.g.
Andalusia is a poorer region than Catalonia), Belgium (where Flanders is more prosperous than Wallonia)
and Germany, a federal state that only was reunited in 1990. Comparing the whole of Europe, inequalities
tend to be the strongest in the former communist countries (Estonia, Bulgaria, Czech Republic).
Table 3.21: Country variation in relation to changes in inequalities
Changes in GDP for OECD countries 2000-2007 (year 2000=100)
Ireland
Czech Republic
Greece
Poland
Iceland
Spain
Sweden
United Kingdom
Norway
Austria
Belgium
Denmark
Netherlands
France
Germany
Portugal
Italy
2000
2004
2005
2006
2007
100
123
131
138
145
100
113
120
128
136
100
119
124
129
134
100
112
116
124
132
100
115
123
126
128
100
113
117
122
127
100
110
113
118
122
100
111
113
116
120
100
109
112
114
118
100
105
107
111
115
100
106
108
111
114
100
104
107
111
113
100
105
106
109
113
100
107
108
111
113
100
102
103
106
108
100
104
104
106
107
104
105
107
100
103
Bulgaria, Cyprus, Estonia, Slovenia are not in OECD data
Societal inequalities are the background for any phenomenon which could be interpreted as digital divide.
Since there is evidence that the degree of inequality varies across countries, it can be inferred that there are
also differences with regard to the digital gaps between wealthy and well educated parts of the population on
the one hand, and socially disadvantaged, less educated part of the population on the other hand.
3.6.3
Urbanisation
There are no comparable data. The information below is based on EU Kids Online National reports.
Most European countries have over the last decade(s) firstly experienced a sustained movement of
population from rural areas to urban centres, leading to rapid fringe developments in most cities. In a second
phase, due to the rapid increase of housing costs in the cities, there was a migration movement from these
inner cities to suburban areas, namely the rise of ‘suburbanisation’. Rather sparsely populated countries
such as Iceland, but also the UK, are predominantly urban and suburban. Some – rather small and dense countries such as Estonia or Belgium can be regarded as being suburban on the whole.
Parallel with the migration to cities and later on to suburban regions, employment in the agricultural sector
has decreased substantially in most European countries. Most European countries only have about 5% or
less employment in agricultural sectors (Austria, Belgium, the Czech Republic, Denmark, Estonia, France,
Iceland, Ireland, Italy, Spain, Sweden, the UK). By contrast, eastern European countries still count more
substantially on agriculture, such as Bulgaria (50% of the population are working in agriculture) and Greece
(11%)
108
The degree of urbanisation turns out to be an indicator of internet penetration. Firstly, urban populations in all
European countries tend to be more online than rural populations. Secondly, particularly dense countries
with large urban and/or suburban regions have a high rate of internet penetration. This is especially true for
France (96%), the UK (80%), the Nordic countries and the Netherlands. Regionalised countries with large
autonomous regions/states such as Germany, Spain or Belgium show large differences in internet access
between the different states/regions.
Countries with important rural regions and an important agricultural activity (such as Greece and Bulgaria),
tend to have on the whole smaller percentages of internet penetration. Some of these countries, and others,
show at the same time strong regional differences (e.g. Ireland, Greece, Italy). Yet, in countries such as
Iceland and Estonia where governments have developed incentive schemes, internet availability has risen in
most rural areas during recent years.
Table 3.22: Clusters of countries according to degree of rural population
Group one:
Countries with very small rural population
Group two:
Countries with a larger rural population
Austria
Belgium
Czech Republic
Bulgaria
Cyprus
Greece
Denmark
Estonia
France
Iceland
Ireland
Italy
Spain
Sweden
UK
Source: Based on national reports.
If it is potentially more difficult to diffuse information in rural areas, one hypothesis that countries with larger
rural populations would have less safety awareness
3.6.4
Work and social class
Collecting figures some aspects of work and social class seems to be quite difficult in most countries. Making
European comparisons proves to be even harder. Figures are not systematically collected nor classified.
Classifications do not match. Most countries do not offer precise statistics on the percentages of the
population involved in manual versus non-manual work.
Figures on sectors of employment or in terms of occupation show rather important similarities. Economies in
most European countries are driven today predominantly by the services sectors (60-70% for Ireland,
Austria, Spain) whereas industry still represents around 30% of employment. Agriculture only counts for
about 5% in most European countries. All of these figures tend to show declining class oppositions and the
growth of a middle-class.
It was not possible to build a classification system for EU Kids Online countries given the problematic nature
of the data.
In principle, one hypothesis might have been that there would be more safety awareness in countries where
the middle-class is larger.
109
3.6.5
Free speech and censorship
Freedom of speech and freedom of opinion seems to be a common thread across all EU Kids Online
members, protected by the very Constitution in Belgium, Denmark, Greece, and Spain. In fact, freedom of
speech is guaranteed everywhere across the project partners - within the confines of the law. Among the excommunist countries, Estonia and the Czech Republic are explicitly mentioned as ranking quite high by the
Reporters without Borders on the Press Freedom Index.
National reports form the EU Kids team mentioned certain aspects of regulation in a number of countries. In
Austria there is provision in slander laws, nevertheless; there is no government restrictions on the internet. In
Bulgaria children have unrestricted access to the internet, similar to adults. In Denmark, administrative law
may demand confidentiality from civil servants in a number of cases. In Germany there is public pressure
towards the regulation of certain kinds of media content and self-regulation is practiced). In Greece freedom
of expression on the internet is regulated in the same way as other print and electronic media and is curbed
in cases of pornographic and obscene material, violation of personal information, misleading advertising and
breach of national security. In Greece, freedom of information and expression on the internet is also under
the protection of the European Convention of Human Rights. In Estonia there is no control over what is
available online; only the big national dailies exercise some form of moderation. In Italy cross-media
ownership under the Berlusconi regime has opened up a debate on free speech and freedom of information.
In Norway commercial content directed to children is regulated, and blasphemous and racist expression is
restrained. In Portugal the newly acquired ‘openness’ of the Portuguese media results in a lack of regulation
of internet content. In Spain restrictions to the freedom of expression apply in cases of privacy protection,
protection of minors, respect for the rights of others, libel and reporting of Basque nationalist terrorism. In
Sweden self-censorship applies. In the United Kingdom following 9/11, restrictions on speech that incites
religious hatred have been imposed, giving cause for concern. Censorship is stricter than elsewhere,
particularly when it comes to sexual issues (e.g. the Netherlands the law is more tolerant).
We can see similarities and differences in this detail, but it was not possible to classify EU Kids Online
countries based on this.
3.6.5.4 Hypotheses regarding the influence of this contextual factors on Safer Internet issues
In principle, one might have anticipated that it would be harder to regulate certain online content in countries
where freedom of speech was valued.
3.6.6
Migration and cultural homogeneity
Eurostat/OECD provide statistics on the proportion of inhabitants with migration background etc.
There are clear differences between the countries on the issue of migration and cultural homogeneity. On
the one hand there are countries with a high level of cultural diversity and a mixture of different nations. An
example of this is Belgium where there are effectively at least two nations. On the other hand there are
countries like Iceland where until the 1980s about 99% of the population was of Icelandic origin. A common
theme for many countries is an increased number of non-nationals during the 1980s and 1990s that for many
countries has meant an enhanced sense of multiculturalism. In addition, many countries have seen stricter
immigration policies with prospective citizens being required to take language exams and exams on basic
knowledge (for example, in Germany). But it is not only the number of immigrants or foreign nationals which
matters. It is also important to look at where the immigrants are coming from and to what extent the
dominant population in each country experiences a shared set of values and ideas with members of minority
groups or with immigrants. Two examples are Estonia where there have been tensions between the
Estonian speaking (67%) and the Russian speaking (33%) populations and also Greece where 58% of the
non-nationals come from Albania and are treated with some degree of suspicion.
110
It is interesting to note that most national reports from EU Kids Online teams do not mention anything about
the possible impact of cultural homogeneity on tolerance towards content on the internet. In most cases
those who do come to the conclusion that there is no such effect or at least it is very limited. Most countries
probably have some regulation against hateful material even though this is mentioned by very few reports.
One possibility to take this theme further would be to distinguish between countries where there have been
direct confrontations (not necessarily violent) between different groups. Examples would be Estonia (tension
between Estonian and Russian speaking) and Slovenia (tensions with Croatia and Serbia) and look at if this
has any impact on the on-line culture.
We can see similarities and differences in this detail, but it was not possible to classify EU Kids Online
countries based on this.
In principle, there could be two opposing hypotheses: where migration creates tensions, there is less
tolerance of certain online content vs. where migration leads to multiculturalism, there is more tolerance of
certain online content.
3.6.7
Role of the state
When dealing with the issue of the extent to which the state can be considered to be interventionist or
laissez-faire, most national reports have a similar story to tell - namely that the past years have seen a
development away from heavy state regulation following the new liberalist wave of the 80s and 90s. Looking
at the text in the national reports, eight countries classify themselves as rather or somewhat interventionist.
These are the Czech Republic, France, Germany, Greece, Italy, Iceland and the UK. Only three, Bulgaria,
Cyprus and Estonia, mention a very liberal attitude. However 10 out of 21 countries either do not mention
this issue or do not deal with it directly. It is also worth considering to what extent difference should be
expected between the EU Kids countries in this respect. If the participating countries are marked out on the
Inglehart-Welzel values map we can see that 15 out of 20 countries (Cyprus is not reported on the map) end
up in the upper left corner of the Inglehart-Welzel map54. Indicating that in these countries ideas are bent
towards self expression rather than survival and towards secular/rational values rather than traditional
values.
54
The map is available from: http://margaux.grandvinum.se/SebTest/wvs/articles/folder_published/article_base_54.
111
Figure 3.1: Inglehart-Welzel map of political attitudes
To what extent is the state and/or the government regarded as being responsible for Internet safety
(compared with industry, school, parents)?
When it comes to how the state acts towards the internet it is firstly important to distinguish between
countries with high internet diffusion and those with low internet diffusion. In countries with low diffusion the
question of the responsibility of the state is at least partially focused on the issue of connectivity. Leaving the
question of connectivity aside, however, there seems to be a distinction between the countries on at least
two issues. Firstly, there is the issue of regulation. On this issue there is a difference between countries
where the emphasis is on centralised regulation and countries where the emphasis is on diffused regulation
or self-regulation. But centralised regulation is not always pursued through government. It can also be
pursued through the dominant service provider such as the former public telecommunications company.
Bulgaria, Czech Republic, Estonia, Iceland, Ireland, Italy, Portugal and Sweden all mention an emphasis on
self regulation whereas Cyprus, France, Germany, Greece, Spain and the UK talk about an emphasis on
state regulation.
Another important issue is education where Sweden and Iceland, for example, talk about an emphasis on
teaching children safe use of the internet. Based on this it seems possible to suggest a two dimensional
classification of the countries into countries where there are different regulatory approaches and countries
where there is a different level of emphasis put on educating children in “proper use” of the internet.
A number of hypotheses are possible. One is that the nature of regulation makes a difference: for example,
is there more safety awareness in countries with more centralised regulation? Another hypothesis would be
that there is more safety awareness in countries where an educational approach is stressed.
3.6.8
Language
For a systematic comparison the Eurobarometer 2006 “Europeans and their languages” can be used. And
Eurydice provides data on English lessons at school.
112
Overall, English appears to be the common language of communicative used over the internet, apart from
the mother tongue, that is, in which there is online content in all project member countries. English is also
almost everyone’s first foreign language taught at school as part of the curriculum from quite early on. In
some countries, such as Greece and Slovenia, English is also taught in special language schools. In most
cases where English is a foreign language, younger generations are more proficient and skilful in it in
comparison to older ones, and can subsequently browse the global internet (in English) more aptly. Some
countries are renowned for their inadequacy in English language skills, i.e. Austria, Germany (there is
considerable internet content available in German catering for audiences in Austria, Germany and
Switzerland), the Czech Republic, France, Italy, and Spain, while in others, the majority of the population is
English-literate and proficient, i.e. Belgium, Denmark, Greece, Iceland, Norway, Sweden and the
Netherlands. Arguably, the following countries can be seen to be somewhere in the middle regarding their
English language skills: Bulgaria, Estonia, Slovenia.
In a few countries, there is online content in languages other than the national idiom and English, such as in
Belgium (officially trilingual); in Iceland (Danish and other Scandinavian languages) and in Estonia (Russian).
Table 3.23: Countries organised by their relative literacy in English
Group one:
Relatively more English literate
Belgium
Denmark
Greece
Iceland
Norway
Sweden
The Netherlands
Ireland (First language)
UK (First language)
Group two:
Relatively less English literate
Austria
Czech Republic
France
Germany
Italy
Portugal
Spain
In countries with more English literacy there would be more concern about what can be accessed online
because the English-speaking world wide web is larger.
3.6.9
‘Bedroom culture’
There are no comparable data
In Bulgaria, families who can afford to have a home computer and a home internet connection are the
minority, hence it is safe to say that there no substantial bedroom culture in Bulgaria. In Estonia, due to
shortage of private space, the computer is usually in a common area and presumably under parental
supervision. Parents, however, are not aware of internet-related risks. When outdoors, children are
supervised until the age of 11-13, though restrictions are more lax in the countryside, which is considered
safer in comparison to the urban milieu. Interestingly, bedroom culture was felt to be not so prevalent in
France. Austria demonstrates media rich bedrooms, with 37% of 6-10 year-olds and 53% of 11-14 year-olds
owning a TV set; 30% of children have a computer in their bedrooms.
Denmark has a high level of media access in children’s bedrooms. Based on the National Study of Danish
cultural and leisure time habits, in 2004 69% of children between 7 and 15 had television in their bedroom,
80% had radio, 91% had stereos, 45% had videos/DVD players, 39% had a computer and 42 % had a play
station, x-box or other game computers, 20% had internet access in their bedroom and 59% have their own
mobile. Most children however are very active offline (and outside bedrooms) as well as online, participating
in sports activities, music, etc.
113
In Germany, the majority of children aged 8-11 have their own room, however, 73% of them spend their
leisure in facilities such as sports clubs. Children from an ethnic background and from a lower social class
are half as likely to be in clubs compared to those from higher social classes.
In Greece, emergent concerns about playing outdoors lead to increased parental supervision of children’s
activities and a bedroom culture- especially in urban areas. When it comes to risks, Greece ranks rather low
in most of the Unicef child well-being index (youngsters having been drunk twice or more, having smoked
cannabis and cigarettes), though about one fourth of Greek teenagers have been bullied.
In Iceland and Norway the weather calls for an indoor culture, but most children live in a media rich
environment. In Sweden, domestic space has increased since the 70s which led to more children having
their own bedrooms. Children live in media rich homes, inheriting the older versions of their parents’
technologies. Owning a PC is thought to be a good thing for a child.
In Italy a bedroom culture is slowly emerging where gaming occupies a significant place. At the same time,
Italian teenagers take part in various activities outside the home, such as sports and cinema. Overall, there
is no parental control of children’s internet access.
In Ireland, the lack of appropriate leisure facilities for children has led to the National Play Policy and a
discussion towards safe public play spaces for children. Children’s first preference is outdoor play but as
they get older they tend to opt for their bedrooms.
In Spain, almost half of PCs owned by children are in children’s bedrooms- which are often shared however
with other siblings. Almost 60% of children owned a mobile in 2006, while 31% of TV sets can be found in
children’s bedrooms- otherwise in a common area.
In the Netherlands the danger of traffic has driven a lot of children back to there rooms. They play out less
than previous generations. Also the diffusion of audio-visual devices makes the own room more attractive
than before.
In the UK, growing affluence over the past decades, the lack of leisure activities for children and youth
outside the home, and growing concern about children safety in public spaces have led to a media rich home
environment and a bedroom culture. In terms of (Unicef) risks, the UK comes on top (or close to) on a
number of these, such as having consumed alcohol twice or more aged 11, having used cannabis aged 15,
having had sex aged 15 and having been bullied.
Table 3.23. Countries organised by the degree to which bedroom culture is widespread
Group one:
Widespread bedroom culture
Austria
Denmark
Germany
Iceland
Ireland
Italy
Norway
Spain
Sweden
The Netherlands
The UK
Group two:
No widespread bedroom culture
Bulgaria
Estonia
France
3.6.9.4 Hypotheses regarding the influence of this contextual factors on Safer Internet issues
One hypothesis would be that in countries where bedroom culture was more widespread it would more
difficult for parents to supervise what is happening in the privacy of these rooms. One caveat is that in
countries where it is not widespread, if children access the internet in spaces outside the home it can also be
difficult for parents to monitor what they are doing.
114
4. Conclusions
4.1
Overview
This report has sought to identify and explain the pattern of cross-national similarities and
differences in children’s online use, skills, opportunities, risks and safety. To do so, it has drawn on a
sizable evidence base in Europe, collated across 21 countries.
This report has argued that, without a comparative perspective, national studies risk two fallacies – that of
assuming one’s own country is unique when it is not, and that of assuming one’s own country is like others
when it is not. However, with a comparative perspective, it is easy to become overwhelmed by both the
volume of data and its many complexities and limitations (a problem for the identification of findings) and by
the multidimensional diversity of social, economic and cultural factors that differentiate the countries within
which such data has been generated (a problem for the explanation of findings).
To identify and explain the available findings, we have produced a theoretical framework (see figure
1.1, ch. 1) that specifies hypothesised relations among key variables as follows:
!
Having access to and making use of the internet is a prerequisite for encountering both opportunities
and risks online.
!
The development of attitudes towards and skills in using the internet both depends on and stimulates
further access and use.
!
Each of these factors influences – facilitating or reducing – the experience of online opportunities and
online risks.
!
All of these above factors is expected to vary according to the age, gender and socioeconomic status of
the individual child.
!
These factors and their interrelations should be understood in social terms – the child is always
embedded in a social context, and parents, teachers and peers are especially likely to mediate their
online experience.
!
While the above relations are, broadly speaking taken to apply universally (i.e. to be cross-national
similarities), it is highly likely that the values on each factor (e.g. amount of use, role of gender, degree or
type of risk, nature of parental mediation, etc) will vary by country.
!
At a country level, such cross-national differences may be explained by any of numerous contextual
factors, particularly including the media environment, ICT regulation, public discourses, cultural attitudes
and values and the educational system of a country.
In what follows, we summarise the main findings and conclusions of our comparative analysis,
focusing on how they support, qualify or contest this framework. The conclusions fall into four parts.
1. The overall pan-European similarities that have been identified are summarised, focusing on the
individual level of analysis.
2. The classification of countries in terms of children’s online risk is reiterated, focusing on the country
level of analysis.
3. Most crucially, the contextual factors are examined for their potential in explaining the country
classification.
4. Conclusions are drawn regarding the effectiveness and limitations of the comparative strategy
employed.
115
4.2
Summary of findings from the comparative analysis
In section 2.5.1, a series of key research questions and hypotheses were examined in relation to the
available findings across Europe. These permit some general conclusions (i.e. cross-national similarities)
which hold, with exceptions, and notwithstanding the limitations on data quality, across the European
countries examined. These are summarised as follows:
Online access and use
!
The evidence across Europe shows that, notwithstanding considerable cross-national differences in
children’s internet use (see next section), the more parents use the internet, the more children do so
also. This applies at both a national level (i.e. countries where parents are more likely to use the internet
are also countries where children are more likely to use it) and at an individual level (i.e. if an individual
parent uses the internet, especially at home, they are more likely to have a child who uses it). It was
concluded that parents use the internet both in order to encourage their children and because they have
been encouraged to do so by their children.
!
Contrary to the widespread assumption that, in general, children are the digital natives and parents the
digital immigrants, it seems that (a) although children (under 18 years) use the internet more than adults
in general, they use it less than parents in particular, and (b) this is particularly the case for those under
11 years but (c) those aged 12-17 are more likely to use the internet than are parents (87% vs. 65%). It
is teenagers, therefore, who are the digital pioneers in Europe.
!
These findings suggest that, in general, for younger children, it is reasonable to expect that their
parents will understand the internet sufficiently to guide their use, but this may not hold for teenagers.
Further, even though internet use may be low among the adult population, it is more likely that parents
will be sufficiently familiar to undertake a mediating role with their children.
!
Across Europe, children are equally likely to use the internet at home and at school, and there is a
positive correlation between use at home and school across countries. The more children use the
internet at home in a country, the more they are likely to use it also at school, and vice versa.
Online risks and opportunities
!
Across Europe, a fair body of research evidence suggests that adults and children agree that children
use the internet as an educational resource, for entertainment, games and fun, for searching for
global information and for social networking, sharing experiences with distant others. Other opportunities
(e.g. user-generated content creation or concrete forms of civic participation), are less common.
!
These opportunities were classified into 12 cells according to the motives of those providing online
contents and services and the relation of the child (as recipient, participant or actor) to that provision.
However, there is little cross-nationally comparable evidence regarding the incidence and take-up of
these various opportunities and, consequently, little can be said regarding the possibility of crossnational differences in online opportunities.
!
It was further proposed that each child climbs a ‘ladder of online opportunities’, beginning with
information-seeking, progressing through games and communication, taking on more interactive forms of
communication and culminating in creative and civic activities. Though many variants are possible, one
implication is that communication and games playing may not be ‘time-wasting’ but, instead, a
motivational step on the way to ‘approved’ activities.
!
Although risks are particularly difficult to define in culturally-consensual ways, and they are difficult to
research in methodologically-rigorous and ethically-responsible ways, a classification of 12 categories of
risk was proposed as likely to be relevant across Europe (and beyond).
!
In terms of overall incidence, these findings provide the basis for a tentative country classification
according to likelihood of encountering online risks (next section). Some cross-national similarities
can also be discerned, particularly in terms of the rank ordering of risks in terms of likelihood.
116
!
Thus, across Europe, notwithstanding considerable cross-national variation, it appears that giving out
personal information is the most common risk (approximately half of online teenagers), that seeing
pornography is the second most common risk at around 4 in 10 across Europe, that seeing violent or
hateful content is third most common risk (at approx one third of teens), that being
bullied/harassed/stalked affects around 1 in 5 or 6 teens online, that receiving unwanted sexual
comments is experienced by between 1 in 10 teens (Germany, Ireland, Portugal) but closer to 1 in 3 or
4 teens in Iceland, Norway, UK and Sweden, rising 1 in 2 in Poland. Last, as regards meeting an online
contact offline, this is the least common but arguably most dangerous risk, showing considerable
consistency in the figures across Europe at around 9% (1 in 11) online teens going to such meetings,
rising to 1 in 5 in Poland, Sweden and the Czech Republic.
!
Several risks are yet to be researched comparatively – self harm, race hate, commercial exploitation.
!
In several countries, a degree of distress or feeling uncomfortable or threatened was reported by
15%-20% of online teens, suggesting, perhaps, the proportion for whom risk poses a degree of harm.
!
Some of the high reports of risk – in Estonia, Poland, Czech Republic – require urgent awarenessraising. Similarly, the advent of new forms of online activity – e.g. social networking – points to the
need for urgent new advice to children and young people. As estimates for now-familiar risks continue to
be substantial, these too require continued attention to keep them in children’s minds.
!
Findings from the pan-European Eurobarometer survey suggest that, according to their parents,
children encounter more online risk through home than school use (though this may be because
parents know little of their children’s use at school).
!
However, among those children who use the internet in an internet café or at a friend’s house, these are
also risky locations, according to parents (especially compared with school use).
!
Complicating policy interventions regarding online risk, it was suggested that increasing opportunities
tends to increase risks, while decreasing risks tends to decrease opportunities. This suggestion
remains for further research to support or contradict.
Online attitudes and skills
!
Overall, the evidence supports the hypothesis that internet-related skills increase with age. This is
likely to include their abilities to protect themselves from online risks although, perhaps surprisingly, this
has been little examined.
!
Boys often claim higher skill levels than girls, though this remains to be tested objectively.
!
Across countries, those in which a higher percentage of parents claim their children have encountered
harmful content tend also to be those in which parents estimate their children to have a lower ability to
cope with these potentially harmful encounters. This negative correlation at the European level clearly
indicates cross-national differences, though the interpretation is as yet unclear. Note that this correlation
does not hold at an individual level (i.e. it cannot be said that if a parent claims their child has
encountered harmful content, that parent is also more likely to think their child can cope).
!
Indeed, though there is growing evidence of the array of coping strategies children employ when faced
with online risk, these are not yet systematically studied and nor is their effectiveness evaluated.
!
There are difficulties measuring internet-related skills as yet, and little available comparable research on
children’s attitudes to the internet.
Age, gender and socioeconomic status
!
Use of the internet increases with age, at least up until the early to mid teens, when usage may peak.
While this trend holds across Europe, in high use countries, children get online younger, and this has
implications for risk – notable since high risk countries (see later) include low and high use countries.
117
!
Generally, it seems that older teenagers encounter more online risks than younger children, though
the question of how younger children cope with online risk remains little researched.
!
The findings also suggest that boys use the internet for longer and in more places than girls do, and
that girls and boys differ in the online activities they engage in - girls prefer activities that involve
communication, content creation and collaboration, boys prefer competition, consumption and action.
!
There are also gender differences in risk: boys appear more likely to seek out offensive or violent
content, to access pornographic content or be sent links to pornographic websites, to meet somebody
offline that they have met online and to give out personal information; girls appear more likely to be
upset by offensive, violent and pornographic material, to chat online with strangers, to receive unwanted
sexual comments and to be asked for personal information but to be wary of providing it to strangers;
both boys and girls are at risk of online harassment and bullying.
!
In almost all countries, higher SES households are more likely to provide their children with access to the
internet, this resulting in greater or more frequent use among more advantaged children. It also appears
that lower class children are more exposed to risk online.
Parental mediation of children’s online activities
!
Parents practice a range of strategies for mediating their children’s online activities - they favour
time restrictions, sitting with their children as they go online and discussing internet use, tending to prefer
these social strategies to technical mediation (filtering, monitoring software). However, there are
differences cross-nationally in preferred strategy that invite further analysis.
!
More consistent across Europe is the tendency for higher SES parents to mediate their children’s
internet use, and for girls to be more subject to such mediation than boys.
!
With regard to age, the consistent finding is that of a U-curve: that parental mediation increases with
age until the age of around 10-11 years and then decreases again.
!
It is unclear, on the present state of knowledge, that any of these strategies is particularly effective in
reducing children’s exposure to risk or increasing their resilience to cope.
4.3
Classification of countries in terms of children’s online risk
In section 2.5.2, the differences identified across countries were used to construct a classification of
countries in terms of children’s online use and risk. Specifically:
!
Although generally European children are gaining access to the internet, differences in access and
use remain, enabling a country classification based on the percentage of children who use the internet.
!
Also striking is the diversity of online risk figures obtained across countries, suggesting a classification
of countries based on the likelihood of children’s experiencing online risk.
!
Putting these two classifications together produced the following table:
Children’s internet use
Online risk
Low
Medium
High
Low
Cyprus
Italy
Greece
Portugal
Spain
Medium
France
Germany
Austria
Ireland
High
Bulgaria
Czech R.
Poland
Slovenia
Estonia
Netherlands
Norway
UK
118
Belgium
Denmark
Sweden
!
As noted earlier, this suggests that: (i) high use of the internet is rarely if ever associated with low risk;
(ii) low use of the internet may be associated with high risk but not vice versa; (iii) high use, high risk
countries are, for the most part, wealthy Northern European countries; (iv) medium use, high risk
situations are characteristic of new entrants to the EC; and (v) Southern European countries tend to be
relatively lower in risk, though there are differences among them.
!
Putting this another way around, we might conclude that, as a broad generality, (i) Northern European
countries tend to be “high use, high risk”; (ii) Southern European countries tend to be “low use, variable
risk”, and (iii) Eastern European countries can be characterised as “new use, new risk”.
!
There are other country classifications possible, as discussed in this report, particularly that based on
children’s perceived ability to cope with online risk (as reported by parents in different countries) – high
ability to cope is claimed for children in Austria, Belgium, Cyprus, Denmark, France, Germany, and the
UK; low ability to cope is claimed in Bulgaria, Estonia, Greece, Portugal and Spain (intermediate
countries are Czech Republic, Ireland, Poland, Slovenia and Sweden). Across countries, findings for
coping are negatively correlated with parents’ perception that their child has encountered harmful
content on the internet, indicating that high risk countries tend to have low perceived coping skills and
vice versa.
!
Also presented earlier is a country classification based on parental mediation. Here it was shown (table
2.19) that, on the assumption that the degree of television mediation practiced reveals parents’
willingness to mediate domestic media, countries differed in their relative mediation of television and the
internet thus. In Austria, Italy, Poland, Portugal, Slovenia and Spain parents of internet users set rules
for television more than they do for the internet. In Denmark, Estonia, Netherlands and Sweden, parents
set more rules for the internet than for television. In Belgium, Germany, Greece, Ireland and the UK,
parental rules are more or less equivalent. In short, in high use countries, parents mediate the internet
more than they do television. In low use countries, by contrast, they are more likely to mediate television
– suggesting a regulation gap in low use countries (i.e. parents are evidently willing to mediate, since
they do so for television, but lack either awareness or skills to mediate the internet to a similar degree).
!
Various other forms of country differences were noted in chapter 2. This included the finding that in
Poland and Portugal, children between 0 and 17 years use the internet more than parents (i.e. even
younger children are digital natives compared with parents and parents may be thus less able to
supervise their children’s internet use); that in Italy, parents are especially behind their children; that
there is some evidence that in countries with low public or domestic access, children are relatively more
likely to go to internet cafés (e.g. Bulgaria, Poland), and that in Ireland, Netherlands, Portugal and the
UK, girls use internet more than boys.
To the extent that we find cross-national differences rather than similarities, we must turn to the country level
to explain these differences. It can be immediately seen that one simple explanation – country size – plays
little relation, though it is equally likely that a country’s wealth (GDP) is related to internet use. Hence, in
chapter 3, we reviewed the available evidence for six dimensions on which national contexts might vary in
ways that shape children’s online experiences in those countries.
It should also be noted, on the other hand, that the contextual factors identified in what follows appear not to
shape the above-noted pan-European similarities – i.e. to the extent that children’s online experience is
similar across countries, we do not need to examine cross-national differences in context for such factors
appear inconsequential.
4.4
Contextual explanations for cross-national differences
The general model of the research field (see figure 1.1, ch. 1) hypothesises that contextual factors at the
country level, discussed in chapter 3, will influence children’s patterns of online use, opportunities and risks.
As a final step of our comparative analysis, we conceptualized countries as units of analysis in order to
explain, if possible, cross-national differences in children’s online experiences in terms of crossnational differences in these contextual factors.
Given the lack of truly comparable data, this step is particularly challenging since both sides of the argument
- the “dependent” as well as the ”independent” variable - had to be constructed using quite different kinds of
119
empirical data and relying on an on-going process of communicative validation within the EU Kids Online
network. Thus, the following interpretations should be treated as highly tentative. Nevertheless, we believe
that indications of significant relationships between contextual factors and patterns of online behaviour can
provide a strong steer for future policy and research recommendations.
In summary, the discussion in chapter 3 of relevant contextual factors revealed the following hypotheses and
observations.
Media environment
!
Diffusion of the internet in different countries strongly influences children’s use of the internet.
Differences in access and use across European countries are still large. As a consequence, for children
in countries in which internet diffusion has reached an advanced stage, online services are a normal part
of their media environment and everyday life, whereas for children in other countries, internet use
remains something that takes a specific effort or requires particular resources not available to all.
!
Diffusion of the internet not only directly affects children’s access and use but also indirectly influences
the range of online activities, parental mediation and, as a result, online-related risks and opportunities.
One important finding is that gender and SES differences appear to be decreasing in the course of the
diffusion process.
!
Due to the lack of comparable data on safety awareness in the different countries the influence on ISP’s
activities in safeguarding online safety cannot be examined directly. Although it is highly plausible that
safety information provided by ISPs can raise awareness and reduce risks, there is little empirical
evaluation available so far. In this respect there is a particularly urgent need for additional research as
we cannot determine, at present, whether variation in national safety awareness activities accounts for
cross-national variation in use, risk or coping with risk.
!
Interpreting the evidence from national reports it may be assumed that the presence of a strong public
service broadcaster as a (major) content provider for children, offline as well as online, can play an
important role in guiding and teaching children how to use the Internet in a safe and constructive way.
Although this assumption is highly plausible, it is surprising that there is almost no empirical evidence –
even on the national level – evaluating the effects of dedicated online content, which sets out to support
children in using the opportunities and avoiding the risks of the internet. In this respect, there is a
particular need for additional research.
In conclusion, cross-national variation in the amount of children’s use of the internet, which depends
in many ways on cross-national variation in internet diffusion, is a crucial dimension in influencing
children’s experience of the internet in Europe. This is likely to have major consequences for their
online opportunities. However, as noted above, higher use is associated with higher risk, but not
exclusively so – there are also some medium use, high risk countries and at least one low use, high
risk country (Bulgaria). Note that in Bulgaria, as observed in chapter 3, awareness of safety
measures is low and filtering is not popular.
ICT regulation
!
The classification provided by the World Economic Forum indicates that while about half of the countries
judge that they have adequate regulation on internet issues in general, there are still exceptions – such
as Cyprus, Poland and Greece - where more regulatory mechanisms are needed. This seems to
correlate with other classifications fairly well – particularly with general internet diffusion. In short, the
more internet users, the more legislation regulating activities on the internet.
!
It was noted that Anglo-Saxon, Northern and Central European countries have a greater tradition of self
regulation than Latin and Southern European countries, in which legislation plays a more important role
than self-regulation.
!
It also seems that where the internet is less common, more efforts are made in promotion of Internet
use, while once the internet becomes more common, risk awareness and then literacy initiatives gain
priority on the policy agenda.
120
In conclusion, although there appears to be considerable variation in ISP’s activity in safeguarding
online safety, this cannot be straightforwardly related to cross-national variation in children’s use or
risk. Nor, from table 3.6, can one discern a straightforward relation between the development of a
regulatory framework and children’s experiences online, though it is suggested that more developed
frameworks are to be found in countries where internet use is relatively high. Compounding the
challenges ahead, it will be observed that relatively low engagement of NGOs with internet safety
issues was found in several high risk countries (table 3.9).
Public discourse
!
Grouping countries on the basis of media coverage on online risks and opportunities does not lead to
clear patterns. There are countries from north and south Europe that are high or low by some criterion of
coverage. The same is true for internet penetration e.g. the UK and Denmark are in the same group as
Portugal and Greece at one point. Media coverage must be driven by other factors.
!
One possibility is that there are common patterns of conceptions of childhood that lie behind and are
embedded in particular national media coverage. For example, in Norway there is a notion of a ‘natural
childhood’, where sexuality is less of a risk while at the same time discussions of children’s rights is
strong. Such underlying conceptions may well help to shape the nature of how media engage in the topic
of children and the internet.
!
In all the countries what was common was the newsworthiness of risks compared to opportunities –
in all countries over half of all articles reported solely risks, the average of all these countries being
nearly two-thirds. In contrast, at most only a quarter of the media articles covered solely opportunities in
any country and the average was less than a fifth.
!
Looking at different types of risk (content, contact, conduct) different national media have very varied
levels of coverage of the three types of risk. Countries low on content risks like Italy, can be high on
conduct risks, and vice versa if we look at Denmark for conduct vs. contact. Or some countries can be
high or low for some risks, but be medium for others. Hence, media coverage in different countries is
sensitising people to different kinds of risk, which may have a bearing on the degree to which people in
different countries think the various risks are prevalent.
!
One example of striking differences in the relative attention to certain risks is the media coverage of
issues of sexuality, which is mainly coverage of pornography on the net. In some countries this aspect
dominates the risk related media coverage (more than one third of all articles): Belgium, Greece, Spain,
and the UK. In contrast, interest in this issue is shown to be very low in Norway, Estonia and Denmark.
Apart from the influence of particular national histories (e.g. the paedophile cases in Belgium), this
probably reflects different national concerns (at least in the media) about what images of sexuality
children should be exposed to.
!
With regard to the question of the extent to which NGOs shape public discourses the main
commonalities across Europe consist in the respective target groups: in almost all countries NGOs are
focusing on raising the awareness of parents and children and to a lesser extent they target the service
providers. Another commonality is that very few NGOs deal only with safer internet issues. Most of those
working on this topic are NGOs working closely with national child protection agencies and more
generally consist of child protection organisations and some extent parents’ organisations as well.
!
Some countries provided evidence that single media events, e.g. high profile ‘crimes’ or ‘anti-social
behaviour’ have generated intense public discussions, as related to the school killings in Germany and
Finland, a particular posting of youth misbehaving in a school in Slovenia or cases of happy-slapping in
France and Italy. Given the media attention given to these events they might have long-term effects on
the public discourses as they frame the perceptions of journalists as well as of the recipients
In conclusion media coverage on online risks and opportunities varies substantially across Europe.
It may be assumed that parents in the countries with a general high level of risk reporting in the
media (Portugal, the UK and Denmark), will have a higher perception of risks than the average of all
these countries. In countries where press coverage reports considerable concerns about the risks of
121
content online, there will be more parental concern about these issues compared to countries where
that particular reporting is low; the same logic applies to contact and conduct risks.
Attitudes and values
!
The countries can be classified according to the dimensions of individualism and collectivism: 1) UK,
Ireland, Belgium with high/moderate individualism and moderate collectivism; 2) Poland, Bulgaria,
Estonia, Portugal, and Czech Republic with low individualism and moderate collectivism; 3) Austria,
Germany, Slovenia, Spain, Iceland, Italy, France and Greece with moderate individualism and low
collectivism; 4) Denmark, Sweden, and the Netherlands with high individualism and low collectivism.
!
This classification shows there is a high correspondence between cultural values and the overall
country classification as developed in chapter 2.5.2 based on children’s internet use and the degree of
online risk. Countries of group 4) are high use countries with medium or high risk; countries of group 2)
are medium or low use countries with high risk; countries within group 3) are medium or low use
countries with medium or low risk; and countries within group 1), somewhat overlapping with group 4)
are high (or medium) use countries with high or medium risk.
!
Another correlation can be found for the parents’ rules relating to children’s use of the TV and the
internet. Almost all countries, in which parents put more emphasis on the mediation of TV use, belong to
group 3, which can be called “the Catholic Europe”, whereas all countries in group 4, “the protestant
Europe” clearly apply more rules for online use.
In conclusion, the correspondence between general values and patterns of online use and risks
indicates that online behaviour as well as perceived online risks are related to and shaped by
underlying value orientations, which differ across Europe. This means that awareness programmes
have to consider the cultural specificities of single countries in order to reach their target groups.
Educational system
!
With regard to the general level of education, Southern European countries show considerably higher
rates of only pre-primary and primary education than Northern, Central and Eastern European countries.
However, among the younger generations these differences are going to disappear. So far, cross
country differences in children’s online use can be partly explained by different levels of general
education: the higher general education of a country, the higher its children’s online use.
!
European countries differ in the degree to which differences in education and socio-economic status are
transferred to the children’s generation. Unfortunately there is almost no systematic empirical evidence
on SES related differences in children’s online behaviour, but illustrative observations support the
assumption that in more stratified societies, internet use is particularly shaped by SES differences.
!
The technical infrastructure of schools has been massively increased in the last years throughout
Europe. However, as several national reports point out, internet penetration in schools is not the same
as actual use. Most students cannot use internet at schools without some kind of control by adults. Even
in tertiary education, access is not completely without restrictions.
!
In most European countries ICT learning is part of the curriculum (both in primary and secondary levels
of education). In most countries ICT learning constitutes an autonomous subject. Only in a few countries
it is just a cross-curricular subject.
In conclusion, the educational system is a relevant contextual factor for children’s internet use.
Although the evidence available does not allow for systematically checking the hypothesis, it may be
assumed that higher education will help a) children to develop online skills and b) parents to develop
skills in mediating their children’s online use. The technical infrastructure of schools as well as the
way how the internet is integrated in curricula and everyday teaching practices will influence
children’s online use at schools. Since online use at schools is often restricted risks as well as
opportunities are reduced in that setting.
122
Background factors
!
The EU Kids Online network generated a wide range of hypotheses regarding the cultural, socioeconomic and technical factors that might influence children’s online access, use and safety. In many
cases, however, there was too little available comparable evidence to permit examining these
hypotheses.
!
It was possible to classify countries according to their active endorsement of the information society
discourse, but it seemed that this did not relate strongly to country classifications based on risk, coping
or parental mediation, though they are loosely related to the classification based on use. Unsurprisingly,
high use countries are more likely to consider themselves on ‘the leading edge’ in the information
society. Whether this results in higher safety awareness among children and parents is unclear from the
available evidence.
!
Urbanisation may shape children’s encounters with the internet and risk. Countries with large rural
populations (Bulgaria, Cyprus, Greece) are also low use countries. Although it is widely held that
socioeconomic status inequalities also shape children’s access to the internet, we found little use of
comparable indicators applied to children in Europe.
!
In terms of the role of the State, those countries that classified themselves as relatively
interventionist (see 3.6.7.2) tended to be low to medium on both use and risk (with the exception of
the Czech Republic and the UK – medium or high use respectively, and both high risk). Notably, two
countries described as taking a liberal approach (Bulgaria, Estonia) appear to be high risk for children
online.
!
Language – we have noted that English language proficiency tends to be higher in Belgium, Denmark,
Greece, Iceland, Norway, Sweden and the Netherlands and to be relatively low in Austria, Germany, the
Czech Republic, France, Italy, Portugal and Spain. Setting aside the exception of the Czech Republic
(where risk was more contact than content risks) and Spain, one may note that, as hypothesised, the
former group are generally higher on online risk indicators than the latter. Access to English language
content may bring risks as well as opportunities.
!
The personalisation of children’s media (e.g. via bedroom culture) may be influencing children’s
lifestyles but seems to have little influence on their online use or risk. This may change as internet
access in the child’s bedroom (or on their mobile phone) – beyond parental supervision – becomes more
widespread.
In conclusion, the adoption of an information society discourse, plus such socio-structural factors
as degree of urbanisation, may be associated with the degree of internet access and use that
children in different countries enjoy. Other factors appear to be more closely associated with the
degree of online risk encountered – this is seemingly higher where the State is less interventionist in
the regulatory regime, where children are more likely to understand English and, perhaps only in the
future, where personalised internet access is more common.
4.5
Commentary on the comparative process
Working closely together for two years since June 2006, the 21 national teams that comprise the EU Kids
Online network have developed constructive working arrangements designed to capture similarities and
diversity across member states so as to facilitate the identification of common patterns, themes and best
practice. This twin dynamic of recognising difference and drawing out shared understandings was originally
developed in our three-national ‘pilot’ comparison (ref to D3.1) and has proved productive.
Specifically, we developed a comparative strategy in order to ‘add value’ on a European level to the many
national studies conducted in different countries, disciplines and languages (identified in Staksrud et al,
2007). For those in similar or related domains who are contemplating the conduct of an cross-national
analysis of similarities and differences in findings, we propose that our analytic framework and working
methods can be of considerable value.
123
As illustrated in Figure 4.1, below, the strategy required a set of countries (C1… Cn) to work collaboratively
to frame research questions relevant to all (RQ1… RQn). These research questions and hypotheses
provided a means of explicating the possible cross national similarities and differences, trends and
associations that can be derived from the existing research literature and/or are of relevance to safety and
risk policy. These were addressed in turn in chapter 2, as summarised above.
The process of comparative analysis can be represented schematically as a grid.
!
Reading horizontally, country level reports were generated by using the available data to answer each
research questions at the national level (i.e. findings for Belgium, France, UK, etc). These national
reports are available on the EU Kids Online website for the 21 countries included in the network.
!
Reading vertically, comparative reports are generated by using the cross-national data pertinent to each
research question (i.e. findings for age, gender, skills, coping, etc).
!
Insofar as the comparative reports identified cross-national similarities, the focus was on the individual
level of analysis (c.f. Figure 1.1). Insofar as they identified differences, the focus was on the country level
of analysis (i.e. the five contextual factors also shown in Figure 1.1, plus a series of background factors).
Figure 4.1: Overview of the research procedure
This approach, we conclude, achieves a systematic and structured outcome in terms of comparative
analysis. Regarding Kohn’s (1989) main rationales for comparative research, outlined at the start of this
report and here pursued in terms of three of his four approaches, the present strategy permitted us to
achieve the following:
!
Treat countries as objects of analysis in their own right. This approach employs an idiographic lens to
understand countries for their own sake; comparison provides a useful strategy for ‘seeing better’ and
determining what is distinctive (or not) about a country. It was achieved through production of the
country reports.
!
Treat countries as the context for examining general hypotheses. This approach analyses tests general
theoretical models across nations, hypothesising similarities across countries while also permitting
findings of cross-national differences to challenge or limit claims. It was here achieved through
production of the comparative reports at the individual level of analysis.
!
Treat countries as units in a multidimensional analysis. This approach seeks to explain patterns of
similarities and, particularly, differences across countries, by inquiring into the external indicators that
explain how and why nations vary systematically. This was achieved, here through production of
comparative reports at the country level of analysis (i.e. explaining the cross-national classification in
terms of contextual factors).
124
We conclude that this comparative strategy has been broadly successful, and offer the following brief
comments in terms of methodology.
!
Specifically, our approach permitted a clear translation of three main rationales for cross-national
research into an effective strategy for comparing countries on multiple dimensions, as organised through
a clear theoretical framework.
!
The analysis could thereby respect findings of both pan-European similarities and differences. It
could test specific hypotheses and also address open research questions. It could situate each country
in the context of others, and it could situate the individual child in the context of national cultural factors.
!
On the other hand, the process was undoubtedly demanding in terms of research effort – both for
each national research team and in terms of the management of and commitment to a highly
collaborative and iterative working process.
!
The analysis was also limited by the quality and extent of the available evidence base – the many
gaps in the data and the many differences in definitions, sample and methods used for such core issues
as online use and risk meant that all claims and conclusions in this report must be treated as indicative
rather than conclusive.
!
Simply put, some data was weaker than could be wished, some was lacking and some was difficult to
interpret. We proceeded, therefore, on the bold assumption that conducting comparisons is preferable to
saying nothing about pan-European patterns, since some added value must surely be extracted from the
many studies conducted. But we did so with extreme caution, not least in order to stimulate more and
better research in the future.
!
The hardest task, other than locating relevant data and negotiating its significance across the network,
was in producing the country classifications. Some may argue that these are too reductive, turning
differences in degree into absolute differences. But for theoretical and pragmatic reasons, we propose
that country classifications are useful, providing a means of discussing similarities and differences as
well as focusing attention on policy priorities (notably, high risk countries).
!
It is also noteworthy, if unsurprising, that although most available findings were national studies, for
many purposes the comparative European data (mainly Eurobarometer, though other sources were also
useful) provided the strongest basis for cross-national analysis.
!
In terms of quality control, we have sought to explicate the basis for our claims and conclusions
throughout, facilitating a ‘read back’ from conclusions to the evidence base for those and, further back,
to the country reports and original reports of data (available at www.eukidsonline.net) from which they
were derived.
!
Many comparative studies produce the empirical basis for cross-national comparisons but end their work
at the stage of producing a series of country reports, effectively leaving the task of identifying and
explaining observed similarities and differences to the reader. We hope our present work provides a
model for the crucial stage of comparative analysis that can systematize and maximise the benefits of
cross-national research.
!
The EU Kids Online network has a further year of work before producing its final report. In this time, we
will pursue the lines of inquiry established here in two directions. First, we will explore the possibilities of
statistically testing the relations among country-level and individual-level factors through Qualitative
Comparative Analysis techniques. Second, we will draw out the policy implications of the findings
presented here, to form the basis of our policy recommendations, due in June 2009.
!
While it has been our intention to extract as much value for the diversity of studies conducted on topic of
children’s online use, opportunities, risk and safety, there can be little doubt that more research,
rigorously conducted on a strongly comparative basis, is greatly needed.
125
5. Bibliography
Accorsi, L., & Gui, M. (2006). La socializzazione ai nuovi media. Riflessioni sul ruolo della scuola nelle
rappresentazioni di studenti di istituti superiori (Social appropriation of new media. The role of the
school in determining students' representations). Scuola e città, 2.
Anesvad. (2003). Informe Sobre La Pornografia Infantil en Internet Anesvad. Retrieved from:
http://www.anesvad.org/pub/cast/doc/informes/informe.pdf.
Bakardjieva, M. (2005). Internet Society: The internet in everyday life. London: Sage.
Basabe, N., & Ros, M. (2005). Cultural dimensions and social behavior correlates: Individualism-Collectivism
and Power Distance. Revue Internationale de Psychologie Sociale, 18(1), 189-225.
Berker, T., Hartmann, M., Punie, Y., & Ward, K. J. (Eds.). (2006). The domestication of media and
technology. Maidenhead: Open University Press.
Bird, S. E., & Jorgenson, J. (2003). Extending the School Day: Gender, Class and the Incorporation of
Technology in Everyday Life. In M. Consalvo & S. Paasonen (Eds.), Women and Everyday Uses of
the Internet: Agency and Identity (pp. 255-274). New York: Peter Lang.
Bolter, J. D., & Grusin, R. (1999). Remediation: Understanding New Media. Cambridge, Mass: MIT Press.
Bourdieu, P., & Jean-Claude, P. (1964). Les héritiers - les étudiants et la culture. Paris: Les Éditions de
Minuit.
Bourdieu, P., & Jean-Claude, P. (1970). La reproduction. Paris: Les Éditions de Minuit.
Capacent Gallup. (2007). SAFT March - April 2007. A report for the Icelandic parents and teachers
association. Reykjavik: Capacent Gallup.
Cardoso, G., Espanha, R., & Lapa, T. (2007). E-Generation: Os Usos de Media pelas Crianças e Jovens em
Portugal. Lisbon: CIES-ISCTE.
Coleman, J., & Hagell, A. (2007). Against the odds: Risk and resilience in adolescence. Chichester, England:
Wiley.
Currie, C., Gabhainn, S. N., Godeau, E., Roberts, C., Smith, R., Currie, D., et al. (Eds.). (2008). Inequalities
in young people's health: HBSC international report from the 2005/2006 Survey. Copenhagen, WHO
Regional Office for Europe, 2008 (Health Policy for Children and Adolescents, No. 5).
Downey, S., Hayes, N., & O’Neill, B. (2007). Final Report, Play & Technology. Dublin: Office of the Minister
for
Children
and
Youth
Affairs.
Retrieved
from:
http://www.omc.gov.ie/viewdoc.asp?fn=/documents/Research/Play_and_Technology.pdf
EC. (1989). Television Without Frontiers Directive 89/552/EEC of the European Parliament and of the
Council. Amended in the Directive 97/36/EC. European Economic Community.
EC. (1997). Television Without Frontiers Directive 97/36/EC of the European Parliament and of the Council.
Official Journal of the European Communities (Official Journal L 202 of 30 June 1997) 30.07.1997].
EC. (2002a). Framework Directive 2002/21/EC of the European Parliament and of the Council: Official
Journal of the European Communities (Official Journal L 108/33 of 24th April 2002)
EC. (2002b). Authorisation Directive 2002/20/EC of the European Parliament and of the Council: Official
EC. (2002c). Universal Service Directive 2002/22/EC of the European Parliament and of the Council: Official
EC. (2002d). Access Directive 2002/19/EC of the European Parliament and of the Council: Official Journal of
the European Communities (Official Journal L 108/7 of 24th April 2002)
EC. (2002e). Privacy Directive 2002/58EC of the European Parliament and of the Council: Official Journal of
the European Communities (Official Journal L 201/37 of 31 July 2002)
EC. (2006). Eurobarometer survey on Safer Internet. Brussels: Special Eurobarometer 250. Safer Internet,
Brussels.
Retrieved
from:
http://ec.europa.eu/information_society/activities/sip/docs/eurobarometer/eurobarometer_2005_25_
ms.pdf EC. (2007). Audio Media Services Directive 2007/65/EC of the European Parliament and of
the Council Official Journal of the European Communities (Official Journal L 332 of 18th December
2007).
EC. (2007). Eurobarometer on Safer Internet for Children: Qualitative Study 2007. Retrieved from
http://ec.europa.eu/information_society/activities/sip/eurobarometer/index_en.htm.
Eurispes/Telefono Azzurro. (2007). Eight Report on Childhood and Youth. Rome: Eurispes. Retrieved from:
www.eurispes.it
126
Eurydice. (2004). Key Data on Information and Communication Technology in Schools in Europe. Available
at http://www.eurydice.org.
Eurydice. (2005). Key Data on Education in Europe. Brussels: Eurydice. Retrieved from:
www.okm.gov.hu/doc/upload/200601/key_data_2005.pdf.
Eurostudent Report. (2005). Social and Economic Conditions of Student Life in Europe. Federal Ministry of
Education
and
Research.
Retrieved
from:
www.campuseuropae.org/en/support/docs/bologna/eurostudent/eurostudent2005.pdf.
Fluckiger, C. (2007). L'appropriation des TIC par les collégiens dans les sphères familières et scolaires. ENS
de Cachan.
Frydenberg, E., Lewis, R., Kennedy, G., Ardila, R., Frindte, W., & Hannoun, R. (2003). Coping with concerns:
An exploratory comparison of Australian, Columbian, German and Palestinian Adolescents. Journal
of Youth and Adolescence, 32(1), 59-66.
Grimm, P., & Rhein, S. (2007). Slapping, Bullying, Snuffing! Zur Problematik von gewalthaltigen und
pornografischen Videoclips auf Mobiltelefonen von Jugendlichen (Slapping, Bullying, Snuffing! The
problem of violent and pornographic videos on mobile phone of adolescents). Berlin: Vistas.
Haddon, L. (2004). Information and Communication Technologies in Everyday Life: A Concise Introduction
and Research Guide. Oxford: Berg.
Hutchby, I. (2001). Conversation and Technology: From the Telephone to the Internet. London: Polity Press.
IFOP. (2006). Les adolescents et Internet. Liaisons amoureuses... liaisons dangereuses? Report.
Inglehart, R., & Baker, W. (2000). Modernization, Cultural Change, and the Persistence of Traditional
Values. American Sociological Review, 65, 19-51.
Irvine, H., & Williams, P. (2002). Internet use in schools: An investigation into the experiences, abilities and
attitudes of teachers and pupils in junior schools. Aslib Proceedings: New Information Perspectives,
54(5), 317-325.
ITU. (2006). digital.life. ITU Internet Report 2006. Geneva: International Telecommunications Union.
JIM. (2007). Jugend, Information und (Multi-) Media. Basisuntersuchung zum Medienumgang 12- bis 19Jähriger. Stuttgart. : Medienpädagogischer Forschungsverbund Südwest. Retrieved from
http://www.mpfs.de/fileadmin/JIM-pdf06/JIM-Studie_2006.pdf.
KIM. (2006). Kinder + Medien, Computer + Internet. Basisuntersuchung zum Medienumgang 6- bis 13Jähriger. Stuttgart: Medienpädagogischer Forschungsverbund Südwest. Retrieved from:
http://www.mpfs.de/fileadmin/KIM-pdf06/KIM2006.pdf.
Kirwil, L. (2002). Raport z badania FGD: Wzorce korzystania z Internetu przez dzieci w wieku 13-15 lat (
FGD Report on Patterns of the Internet Use in Children Aged 13-15). Warsaw: Research
Department, Ombudsman for Children’s Rights.
Kirwil, L. (2008). Role of Contextual Factor: Individualistic-Collectivistic Value Orientation. EU Kids Online
Network.
Kohn, M. L. (1989). Introduction. In M. L. Kohn (Ed.), Cross-National Research in Sociology. Newbury Park:
Sage.
Lawson, T., & Comber, C. (2000). Censorship, the internet and schools: A new moral panic? The Curriculum
Journal, 11(2), 273-285.
Lee, W.-N., & Choi, S. M. (2005). The role of horizontal and vertical individualism and collectivism in online
consumers' response toward persuasive communication on the Web. Journal of Computer-Mediated
Communication, 11(1), article 15. Retrieved from: http://jcmc.indiana.edu/vol11/issue11/wnlee.html.
Livingstone, S. (2002). Young People and New Media: Childhood and the Changing Media Environment.
London: Sage.
Livingstone, S. (2003). On the challenges of cross-national comparative media research. European Journal
of Communication, 18(4), 477-500.
Livingstone S., & Bober, M. (2004). UK Children Go Online: Surveying the experiences of young people and
their parents. London: London School of Economics and Political Science.
Livingstone, S., & Bovill, M. (2001). Children and Their Changing Media Environment: A Comparative
European Study. Mahwah: Lawrence Erlbaum Associates.
Livingstone, S., & Helsper, E. (2007). Gradations in digital inclusion: Children, young people and the digital
divide. New Media & Society, 9(4), 671-696.
Livingstone, S., & Helsper, E. J. (in press). Parental mediation of children’s internet use. Journal of
Broadcasting and Electronic Media, 52(4), Winter 2008.
Livingstone, S., & Haddon, L. (2008). Risky experiences for European children online: Charting research
strengths and research gaps. Children and Society, 22, 314-323.
Lobe, B., Livingstone, S., & Haddon, L. (Eds.). (2007). Researching Children’s Experiences Online across
Countries: Issues and Problems in Methodology. EU Kids Online Network. Retrieved from:
http://www.lse.ac.uk/collections/EUKidsOnline/Reports/ReportD4.1MethodologicalIssuesCover.pdf.
Martin, O. (2004). L'internet des 10-20 ans. Réseaux, 22(123), 25-58.
127
Martin, O. (2008). La conquête des outils électroniques de l'individualisation chez les 12-22 ans. Réseaux,
25(146), 335-365.
Mediappro. (2006). Mediappro: A European Research Project: The Appropriation of New Media by Youth.
Final Report. Brussels: EC.
Mediaprro. (2006). Mediappro: A European Research Project: The Appropriation of New Media by Youth.
Final Report. Brussels: EC.
Medierådet. (2006). Ungar och medier 2006. Fakta om barns och ungas användning och upplevelser av
medier (Kids and Media 2006. Facts about Children’s and Young People’s Use and Experience of
Media). The Media Council, Ministry of Education and Culture, Sweden. Available at
http://www.medieradet.se/upload/Rapporter_pdf/Ungar_&_Medier_2006.pdf (in Swedish).
Metsoja, A. (2006). Survey results of Estonia. Unpublished MEDIAPPRO Country Report. Tartu: University of
Tartu.
Metton, C. (2006). Devenir grand. Le rôle des technologies de la communication dans la socialisation des
collégiens. EHESS.
MMM Project. (2005). Mina. Maailm. Meedia. (Me. The World. The Media) Tartu: University of Tartu.
Networked Readiness Index (NRI). (2008). Global Information Technology Report 2007-2008 World
Economic Forum. Retrieved from: http://www.insead.edu/v1/gitr/wef/main/home.cfm.
NRI. (2008). Global Information Technology Report 2007-2008 World Economic Forum. Retrieved from:
http://www.insead.edu/v1/gitr/wef/main/home.cfm.
OECD. (2007). Education at a glance. Paris: OECD.
Ofcom. (2006). Media Literacy Audit: Report on media literacy amongst children.
Oswell, D. (1999). The dark side of cyberspace: Internet content regulation and child protection.
Convergence: The Journal of Research into New Media Technologies, 5(4), 42-62.
Pasquier, D. (2005). Cultures lycéennes. La tyrannie de la majorité. Paris: Autrement.
Prensky, M. (2001). Digital natives, digital immigrants. On the Horizon, 9(5).
Rogers, E. M. (1995). Diffusion of Innovations (4th ed.). New York: Basic Books.
Safer Internet plus programme. (2007). Public consultation – Creating a safer online environment for
children.
Information
available
from:
http://ec.europa.eu/information_society/newsroom/cf/itemdetail.cfm?item_id=3355.
SAFT. (2006). SAFT 2006 Parent and Children survey. 2004-2006: Norwegian Action Plan for Children,
Youth and the Internet and the European Commission Safer Internet Action Plan: Norwegian Media
Authority.
Silverstone, R., Hirsch, E., & Morley, D. (1992). Information and Communication Technologies and the Moral
Economy of the Household. In R. Silverstone & E. Hirsch (Eds.), Consuming Technologies: Media
and Information in Domestic Spaces (pp. 115-131). London: Routledge.
Staksrud,
E.
(2005).
SAFT
Project
Final
Report.
Retrieved
from:
http://europa.eu.int/information_society/activities/sip/projects/awareness/saft/index_en.htm#results.
Tezanos, J. F. (2006). Estudio sociológico sobre el uso de Internet en las familias (Sociological study about
the use of the Internet among families). Madrid: Fundación Sistema, Grupo de Estudio sobre
Tendencias Sociales.
Tsaliki, L. (2008). Me, myself and my mobe: mobile telephony in Greece. International Journal of Cultural
Studies (under review), Special issue.
Turu-uuringute
AS.
(2006).
Children
and
Internet.
Available
online
(in
Estonian)
http://www.peremeedia.ee/public/LapsedInternetUuring.ppt.
UK Children Go Online. UK Children Go Online: Emerging Opportunities and Dangers. London: London
School of Economics and Political Science. Project website: http://www.children-go-online.net
Valkenburg, P. M. (2004). Children's Responses to the Screen: A Media Psychological Approach. Mahwah,
NJ: Lawrence Erlbaum.
Ward, K. (2005). Internet consumption in Ireland – Towards a ‘connected’ life. In Media, Technology and
Everyday Life in Europe. In R. Silverstone (Ed.), Media, Technology and Everyday Life in Europe
(pp. 107-123). Ashgate: Aldershot.
Webwise. (2006). Webwise 2006. Survey of children's Use of the Internet - Investigating Online Risk
Behaviour: National Centre for Technology in Education (NCTE).
Wojtasik, ". (2003). Paedophilia and pornography on the Internet: Threats to children. Warsaw: Nobody’s
Children
Foundation.
Retrieved
from:
http://www.canee.net/index/?id=a3f390d88e4c41f2747bfa2f1b5f87db
World Economic Forum. (2007). Executive Opinion Survey 2006 2007. World Economic Forum. Retrieved
from: http://www.insead.edu/v1/gitr/wef/main/analysis/showdatatable.cfm?vno=2.18.
128
6.
Annexes
Annex A: EU Kids Online
European Research on Children’s Safe Use of the
Internet and New Media. See www.eukidsonline.net
EU Kids Online is a thematic network examining European research on cultural, contextual and risk issues in
children's safe use of the internet and new media between 2006 and 2009. It focuses on the intersection of
three domains:
!
!
!
Children (mainly up to 18 years old), their families, domestic users
Online technologies, especially the internet; focussing on use and risk issues
European, cross-national, empirical research and policy
This network is not funded to conduct new empirical research but rather to identify, compare and draw
conclusions from existing and ongoing research across Europe.
It is funded by
the
European Commission’s Safer Internet plus Programme (see
http://europa.eu.int/information_society/activities/sip/index_en.htm) and coordinated by the Department of Media
and Communications at the London School of Economics, guided by an International Advisory Board and
liaison with national policy/NGO advisors.
EU Kids Online includes research teams in 21 member states, selected to span diversity in countries,
academic disciplines and expertise: Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia,
France, Germany, Greece, Iceland, Ireland, Italy, Norway, Poland, Portugal, Slovenia, Spain, Sweden, The
Netherlands and The United Kingdom.
The objectives, to be achieved via seven work packages, are:
!
!
!
!
!
!
!
To identify and evaluate available data on children’s and families’ use of the internet and new online
technologies, noting gaps in the evidence base (WP1)
To understand the research in context and inform the research agenda (WP2)
To compare findings across diverse European countries, so as to identify risks and safety concerns, their
distribution, significance and consequences (WP3)
To understand these risks in the context of the changing media environment, cultural contexts of
childhood and family, and regulatory/policy contexts (WP2&3)
To enhance the understanding of methodological issues and challenges involved in studying children,
online technologies, and cross-national comparisons (WP4)
To develop evidence-based policy recommendations for awareness-raising, media literacy and other
actions to promote safer use of the internet/online technologies (WP5)
To network researchers across Europe to share and compare data, findings, theory, disciplines and
methodological approaches (WP1-7)
Main outputs are available or planned as follows:
!
!
!
!
!
!
!
!
!
Data Repository: a public, searchable resource for empirical research (now online)
Report on Data Availability: a mapping of what is known and not known (Sept 2007)
Preliminary Report Comparing Three Countries (Sept 2007)
Methodological Issues Review (Sept 2007)
Report on Cross-National Comparisons over 18 Countries (Sept 2008)
Best Practice Research Guide (Sept 2008)
Report: Cross-Cultural Contexts of Research (March 2009)
Final Conference (June 2009)
Report: Summary and Recommendations (June 2009)
129
!
Final Report and Book (Sept 2009)
Annex B: EU Kids Online members
Country
Institution
Researchers
Austria
University of Salzburg
Belgium
Catholic University of Leuven
Ingrid Paus-Hasebrink
Christina Ortner
Manfred Rathmoser
Christine Wijnen
Leen D’Haenens
Verónica Donoso
Bieke Zaman
Nico Carpentier
Katia Segers
Joke Bauwens
Jivka Marinova
Mariya Gencheva
Maria Dimitrova
Ilina Dimitrova
Christina Haralanova
Yiannis Laouris
Tatjana Taraszow
Elena Aristodemou
Vaclav Stetka
Jan Miessler
Gitte Stald
Jeppe Jensen
Veronika Kalmus
Pille Pruulmann-Vengerfeldt
Andra Siibak
Pille Runnel
Kadri Ugur
Benoit LeLong
Cédric Fluckiger
Uwe Hasebrink
Claudia Lampert
Liza Tsaliki
Despina Chronaki
Valia Papadimitraki
Thorbjorn Broddason
Gudberg K. Jonsson
Kjartan Olafsson
Brian O’Neill
Helen McQuillan
Simon Grehan
Fausto Colombo
Piermarco Aroldi
Barbara Scifo
Giovanna Mascheroni
Maria Francesca Murru
Ingunn Hagen
Thomas Wold
Elisabeth Staksrud
Petter Bae Brandtzæg
Wies#aw Godzic
Lucyna Kirwil
Barbara Giza
Free University of Brussels
Bulgaria
GERT
Cyprus
Internet Rights Bulgaria Foundation
Cyprus Neuroscience and Technology
Institute
Czech
Republic
Denmark
Masaryk University, Brno
Estonia
University of Tartu
France
France Telecom
Germany
Hans Bredow Institute For Media Research
Greece
London School of Economics
Iceland
University of Iceland
Ireland
University of Akureyri Research Institute
Dublin Institute of Technology
Italy
National Centre for Technology in Education
Catholic University of Milan
Norway
Poland
IT University, Copenhagen
Norwegian University of Science and
Technology, Trondheim
Norwegian Media Authority
SINTEF, Oslo
Warsaw School of Social Psychology
130
Portugal
New University of Lisbon
Slovenia
Lisbon University
University of Ljubljana
Spain
The University of the Basque Country
Sweden
The
Netherlands
University of Gothenburg
The Netherlands Institute of Social Research
The UK
University of Amsterdam
London School of Economics and Political
Science
131
Tomasz "ysakowski
Cristina Ponte
Cátia Candeias
Nelson Vieira
Sofia Viseu
Ema Sofia Leitão
José Alberto Simões
Tomás Patrocínio
Bojana Lobe
Alenka Zavbi
Carmelo Garitaonandia
Maialen Garmendia
Gemma Martinez
Cecilia von Feilitzen
Jos de Haan
Marion Duimel
Patti Valkenburg
Sonia Livingstone
Leslie Haddon
Panayiota Tsatsou
132
Co-funded by the European Union
Contact details
Professor Sonia Livingstone and Dr Leslie Haddon
Department of Media and Communications
The London School of Economics and Political Science,
Houghton Street, London WC2A 2AE, UK
f +44 (0) 20 7955 7248
e [email protected] / [email protected]
The EU Kids Online thematic network has received funding from the European Community’s Safer
Internet Plus programme. The authors are solely responsible for the contents of this report. It does
not represent the opinion of the Community and nor is the Community responsible for any use that
might be made of information contained in it.

EU Kids Online Rep Cover_3_AD.indd

Transkript

Benzer belgeler

The Relationship between Growth and Foreign Trade during the Debt

European identity compared

20 Sep 1977 - 1973 Bitburg Barons

International Summer School